If you’ve been reading my essays here at The Muse, you’ve come to expect a certain level of commentary, opinion, and analysis. This article, which discusses a recent town hall with university students headlined by Senator Bernie Sanders of Vermont and Representative Ro Khanna of California, departs from that style, in that it is more facts than analysis, more “tell me what they said” and less “here’s what I think they meant.” My intent here is to give you as many of the highlights of the event as possible and let the words of the legislators and the students speak for themselves.

But, but… it wouldn’t be a “Muse” piece if I didn’t give it some character. Fortunately, Bernie delivered on that point, just by being Bernie. If you read through to the end, you’ll get two little nuggets that certainly made me smile. Given what’s going on in our world right now, we need all the joy we can get.

On February 20, Senator Bernie Sanders and California Rep. Ro Khanna stopped by Stanford University for a private town hall with the Stanford community. I had been granted a media pass, but the flu slammed me hard against the floor, and I ended up having to watch the livestream in bed. Thank heaven for streaming technology. (Yes, I feel the irony here.)

Organized by the Stanford Speakers Bureau and the Stanford Democrats, the event was titled “Who Controls the Future of Al: THE OLIGARCHS or THE PEOPLE?” Sen. Sanders and Rep. Khanna both delivered brief remarks, which I summarize below, and then opened the floor for student questions.

Bernie Sanders worries about workers

As he typically does, Senator Sanders gave a grounded, nuanced, yet inspiring speech about the exponentially rising power of AI and the self-serving interests of the wealthy corporate owners investing in it. A few highlights, with italics Mr. Sanders’ emphasis, and bolded italics my emphasis:

“I wish that we could be here at happier times, but I would not be honest if I didn’t tell you what you already know, that we are living … in an extremely stressful moment and in many ways, the most dangerous moment in the modern history of this country.”

“We are at the beginning of the most profound technological revolution in world history, a revolution which will bring unimaginable changes to our country and the entire world. These changes will impact our economy. They will threaten our democratic institutions. They will impact our emotional well-being and how we relate to each other as human beings. They will impact how we educate and raise our kids. They will impact foreign policy and the nature of warfare.”

“AI and robotics are creating the most sweeping technological revolution in the history of humanity, and the scale, scope and speed of that change is unprecedented. ... And it’s not just what the AI companies are saying. Take a look at what they are doing this year: four AI companies are spending $670 billion to build out data centers as a percentage of GDP that is 10 times greater than what we spent each year on the moon landing.”

“We are looking at the most significant investment in human history, which is bringing about the most rapid societal transformation in human history. ... The question that we should be asking day after day, but unfortunately we are not ... is, Who is pushing this revolution, who benefits from it, and who gets hurt?”

“At a time in American history when we have more income and wealth inequality than we have ever had, the AI revolution is being pushed by the very, very wealthiest people in our country, including folks like Elon Musk, Jeff Bezos, Larry Ellison, Mark Zuckerberg, and Peter Thiel and others. A small number of multi-multi-billionaires are investing huge sums of money in AI and robotics. And the question that we’ve got to ask ourselves is, why are they doing this?

“Are they doing it because they want to improve the standard of living of the 60% of our people who today in America are living paycheck to paycheck? Many of your families are struggling to pay for health care, pay for housing, pay for the cost of your education, pay for groceries. Stop for a moment and think, is that what Mr. Musk and Mr. Bezos are thinking about, as they generate this extraordinary revolution? Well, some people may think that is the case. I know those guys. I don’t believe it. I think the answer is fairly obvious. The richest people in this country are investing in AI and robotics because those investments will increase their wealth and power exponentially. In other words, the richest and most powerful people on earth will become even richer and more powerful.”

“Jeff Bezos, the fifth wealthiest person in the world, worth about $230 billion, has reportedly pushed his staff for years to think big and envision what it would take Amazon to fully automate its operations. Bezos plans—this is no great secret—to replace at least 600,000 workers at Amazon with robots. According to Mr. Amodei, the head of Anthropic, AI ‘isn’t a substitute for specific human jobs but rather a general labor substitute for humans.’”1 According to OpenAI’s charter, its mission is to build ‘highly autonomous systems that outperform humans at most economically valuable work.’”2

“As the ranking member of the U.S. Senate Health, Education, Labor & Pensions Committee, I released a report finding that AI automation and robotics could replace nearly 100 million jobs in America over the next decade, including 47% of truck drivers, 64% of accountants and 89% of fast-food workers, among many, many other occupations.”

“We are already starting to see the impact of AI on workers and especially young people. Researchers right here at Stanford released the paper in November called Canaries in the Coal Mine that found that there has already been a 16% relative decline in employment for younger workers in jobs exposed to AI like computer programming and customer service. In other words, it is already more difficult for young college graduates to find decent jobs.”

“… what motivated me to get involved in this issue, is my belief ... that the Congress and the American people are very unprepared for the tsunami that is coming. If millions of people lose their jobs and can’t simply go to another job because that other job has also disappeared, what do we do? And somebody says, ‘Well, no problem, we’ll just give basic—a check to everybody.’ Really? Who’s going to write that check? And if they’re going to simply write out checks, how much are you getting? … Is it Mr. Musk who’s gonna be writing out a check?”

“And one of the questions we have to ask ourselves, ‘If AI is going to replace a lot of the work that human beings do, what becomes of human beings? Are we superfluous in the process? What happens to our ability to relate to each other?’ ... if we lose work as an important part of our lives, what do we do with our lives? Not a lot of answers, and people can debate it, but that’s the issue that we have got to begin debating.”

Ro Khanna’s new tech social contract

Ro Khanna spoke next. He called out the fact that we live in a “new Gilded Age,” albeit one for the tech billionaires. These tech billionaires, he said, have told him that they would have been “heroic conquerors” if they lived in a different era. “Musk, Thiel, Ellison, believe they have a modern divine right to lead and rule,” Khanna told the audience. “That’s just not my observation. ... A nation cannot survive with islands of prosperity and seas of despair.”

He cited the work of Professor Gabriel Zucman, who has shown that, in Khanna’s words, “today’s wealth concentration is at the highest it has ever been in our nation’s history, far higher than when I was teaching here just 10 years ago. Listen to this. About 19 billionaires have $3.4 trillion, the equivalent of 12.5% of all the goods and services that are produced in the United States. This is nearly three times more than the wealthiest Americans were worth relative to the size of the economy in the Gilded Age, three times as much wealth concentration today. Extreme wealth often forms an unholy alliance with power to strip us of equal voice and to have two tiers of justice. And that’s really what the Epstein class was: two tiers of justice, rich and powerful people who felt they were above the law and didn’t have to live by the rest of the norms and laws of civilized society.”

Khanna reminded the audience that they’re sitting at the very epicenter of the wealth built up by tech innovation. “This 50-mile radius around my district includes Stanford, Apple, Google, Nvidia, Broadcom, Tesla… [together they’re]3 worth over $18 trillion. We met with some of these companies earlier today. Think about that. One-third of the entire nation’s stock market is in this district. One third!”

The question we need to ask ourselves, said Khanna, is what kind of future we’re building. “Will this future be only for the tech lords, for the tech billionaires, or will it be for all of us?” He underscored a fact that doesn’t get much play in the mainstream media, and that is, that the development of AI has been funded with taxpayer money and philanthropic dollars at academic institutions like Dartmouth, MIT, Stanford and Princeton (e.g., ImageNet and the Stanford Integrated Digital Library Project, funded by the National Science Foundation, in which Larry Page and Sergey Brin participated and which was the precursor of Google.) It’s important to know this history, because, as Khanna said, “just like every other successful generation of American entrepreneurs over the past two centuries, [the AI companies] stand on the foundation of public investment. That is why we must ask not what America can do for Silicon Valley, but what Silicon Valley must do for America.” (emphasis mine)

“Today,” Khanna continued, “I call for a new tech social contract.” Calling for shared prosperity for all, a thriving middle class, and decent jobs in every community, from rural areas and factory towns to suburban neighborhoods and urban centers, he laid out seven principles for what a “democratic AI” should look like:

Principle I: Keep human beings in the loop. Implement real protections against mass displacement. The AI we need is one that “augments human capability instead of eliminating jobs.”

Principle II: Ensure workers benefit from the productivity gains that AI is generating. Every large company must bargain with its workers, and unions or elected representatives should ensure workers move into new high value jobs, or can share in AI’s productivity gains through higher wages, profit sharing, and shorter work weeks. [lots of applause here]

Principle III: Fix the tax code’s “anti-human bias.” Khanna contrasted the “accelerated depreciation” and zero tax companies are granted for using digital tools against the approximately 30% in taxes they pay when they hire human beings. He also called for an annual data dividend, in effect a payment to every American whose data is used by private sector and government, coupled with a “data Bill of Rights” to prevent surveillance companies from sharing our data with deportation and detention agencies. [loud applause here]

Principle IV: Launch a future workforce administration for white collar and blue collar workers and their families, funded by a modest wealth tax on the trillions created here in Silicon Valley, and a token tax on the AI used by businesses that could displace labor. This program can put millions of Americans to work in public service and establish new trade schools and tech institutes to prepare the next generation for careers that AI cannot replace.

Principle V: Data centers must serve the communities that power them. Currently, data centers are “one-way extraction centers from communities to the wealthiest corporations.” Tech companies need to provide compute resources for schools and libraries, create local jobs and use renewable energy and dry cooling technology instead of water. Most importantly, they should pay their own energy bills.

Principle VI: Prevent AI from weaponizing public discourse. We need to stop algorithms from spreading hate and hold platforms responsible for amplifying violent content.

Principle VII: A robust federal agency to regulate AI, similar to regulation of nuclear energy or federal aviation. We need clear, enforceable, non voluntary guardrails and mandatory third-party verification of advanced AI models. Voluntary self-regulation does not work.

Khanna told the audience that neither he nor Sen. Sanders are against innovation, as certain voices loudly claim, but that they simply wish to ensure that the benefits of AI reach everyone, not just the wealthy. “The future must not be written by AI agents that serve San Francisco billionaires,” he said. “It must be written by all of us together in a way that … gives us a new national purpose of economic renewal and independence for every American.”

What do the young people want to know?

After their remarks, Sen. Sanders and Rep. Khanna took their seats on the stage for the Q&A. Two long eager lines of students quickly formed—too long, of course, for everyone to ask their question. Ultimately, only eight of those students (five men and three women) were able to do so because the questions prompted long and thoughtful responses from the legislators, yet one hopes these eight questions were at least to some degree representative of the others that went unasked. The questions ranged from the topic of leadership to innovation to labor to perhaps the most morally striking question about the “rights” of undocumented students to receive education at a place like Stanford.

Make a big difference for your people

Two of the students were first-generation, from low-income backgrounds. One of them, from the Appalachia region in rural West Virginia, mentioned the recent passage of House Bill 2014, which allows high-impact data centers to be constructed in rural communities given that, as he explained, “most of our communities are depressed coal towns, and so we really don’t have an economy. The narrative in my state is that our economy will be replaced by these, quote, data centers.”

The young man seemed rather skeptical about this narrative his state’s legislators were promoting to their constituents. “Do you really think that these new data centers will really be the source that will resurrect the economies of rural America?” he asked the two leaders onstage. “Do you think that working-class Americans like my people will actually lift out of poverty [thanks to these data centers] to revive these depressed coal towns, or are my people being once again tricked by the wealthiest Americans and by the Right in Congress?” Wondering out loud whether West Virginians might possibly be “misunderstanding these data centers and their roles,” he posed the question “what are the pathways [of] messaging, [of] letting my people and my community know the realities and the effects to their everyday lives?”

This elicited an impressed “Wow” and a sobering chuckle from Senator Sanders, who took a few moments to compose his response. First, he said, data centers do bring in a lot of jobs, but those are construction jobs. “The evidence is also pretty clear,” he added, “that once the data centers are up, the number of [local] jobs are fairly minimal.” He then told of conversations with people during his trip to Trump country—a county in West Virginia that had voted for Trump by about 70%. “We have got to understand, where we are politically in America is, in my view, [that] most working class people are not racists; they’re not sexist; they’re not homophobes. Most working people are desperate.” These people, Sen. Sanders explained, have all but given up on the Democratic Party because despite all the rhetoric, they still can’t afford groceries, housing, or health care, and they certainly can’t afford to send their kids to college. “So my hope is,” he told the young man, “if you’re going back to your home state, they are in desperate need of good leadership. Help build a political movement which will represent the great working people of West Virginia.” He added, “Make sure that we have a moratorium on data centers in general.”

For his part, Rep. Khanna seconded Sen. Sanders’ remarks. “It is lazy for political figures to say, ‘I brought in a data center; now we’re going to create new jobs here,’” he said. “We need a far more comprehensive approach. I’ve called for a Marshall Plan for America: advanced investment in manufacturing, in childcare, in elder care, in tech centers, in trade schools—to actually have economic development.” Commending the young man on his love for his community, Khanna encouraged him not to take plumb job offers from New York or San Francisco upon graduation, but instead to “go back to your hometown, go back to West Virginia, if you can, and get involved in politics. I think you’ll make a big difference.”

How galling: undocumented students getting freebies

Another young man stepped up to the mic. “How can you justify a system,” he asked, “where my friend Jay, an international student who has lived in California for over four years, is forced to pay over $80,000 for out of state tuition, while undocumented students qualify for subsidized education through the California Dream Act?”

This of course is the kind of question that places the blame on the poor, the immigrants, the undocumented, rather than taking a closer look at the system that leverages fear and lack to demonize the very people it oppresses. In other words, it’s a blame-the-victim mindset.

“I think that’s a false choice,” Khanna said. “Yes, your friend shouldn’t have to pay $80,000, but we can have free public college in this country if we just had a tax on the Wall Street financial speculation.” He gave another comparison: it would cost about $80 billion to give everyone in the U.S. a free public college education—less than 10% of the defense budget. There are a number of different ways to calculate the cost of free university-level education. According to the Education Data Initiative, as an example, the maximum annual cost of free college would be equivalent to 1.07% of the federal budget ($6.75 trillion for FY2024).

Thanking the student for his question while also making the point that it’s a little misguided, Sanders appealed to the country’s need for more doctors, engineers, and other professionals, to “help transform this country.” He added, “You are living in the richest country in the history of the world. Two years ago, Donald Trump gave a trillion dollars in tax breaks to the 1% and we cut back on education. Does that make sense to you?”

Why is the US such a powerhouse while the EU lags behind?

The last question was one that doesn’t usually see the light of day. “Despite similar levels of education and development, Europe has produced almost none of the world’s largest frontier tech firms, while the United States dominates global tech innovation,” this student began. “I want to ask, what institutional or economic features do you think explain this difference?” He explained he’s not looking into the “why” of this apparent technological inequality between the US and Europe, but rather the “why” of such powerful growth in the US over the last few decades. Without batting an eye, Sen. Sanders responded with a nod to the hard work of America’s entrepreneurs: “Because we are very smart people who have been incentivized, who have worked very hard.” He understood where the student’s question was going—regulation vs economic incentives. If we don’t regulate the companies, he said, and if we give them tax breaks, of course they’ll be able to innovate so much faster and on a larger scale.

But Sanders wasn’t taking the bait. He immediately pivoted to the deeper issue at hand, which no analysis of economic growth or technological innovation should avoid, and that is the value and impact of a technology on society and people’s lives. “These technologies can be very useful,” he said, “[but] they will not be useful unless we have a degree of control over what [the founders] are doing. So it’s not just technology for the sake of technology; it’s technology to improve human life.”

The young man insisted on his original point—the creation of wealth and the rate of growth, regardless of the fact that “we don’t provide for the poor as much as we should.” So Sanders but it a little more directly: “It’s not a question of the poor. 800,000 people are homeless, 60% of our people living paycheck to paycheck in the wealthiest country in the history of the world. So what Ro is saying, and what I am saying, [is] we’re not against AI and we’re not against robotics. But we want those technologies to improve life for all people, and not just make the very wealthiest people in this country even wealthier.”

That got a lot of applause.

Khanna took the mic next and expounded a little more on the causes for America’s faster rate of technological innovation. America used to invest in education, he explained, and that catapulted us forward on the world stage. He cited Gary Becker, the late Nobel Laureate in Economics, who said that in a digital economy the two greatest investments you can make are in the health and education of your people. In post-WWII Europe, he said, high school was the domain of the elite, with just 30-40% of the population having access, while here in the U.S., 80% of our population had a high school education. We “invested in our research universities and ... had a policy of welcoming people from around the world; [it was] immigrants that built our economic growth.” He said he very much appreciates our entrepreneurial culture, but there is something that does frustrate him, and that is when he meets a founder who says “I’m self-made.” In Khanna’s mind, he said, “I don’t think I’m self-made. My grandfather was in jail for four years, under colonialism. Be hard for me to think I’m self-made. You had health care, you had child care. You had parents who encouraged you. So, yeah, okay, you ... didn’t start out with $100 million, but, have the humility to understand what you were given by this country in terms of your start.”

The room broke into applause once again.

How to become president of Poland

The question that made the greatest impression on me personally was that of a young woman from Poland, possibly in part because I myself come from one of its neighboring countries. The young woman shared that her dream has always been to be the “first female president of Poland.” Well, it just so happens that Bernie Sanders’ father also came from Poland, “without a nickel in his pocket,” and his wife’s family came to the U.S. from Ireland. Sanders looked out over the packed house. “Many of your families came from all over the world,” he said. “That’s who we are as a people.” He then exhorted the young woman to work with people at the grassroots level: “Don’t start thinking, ‘I want to be President of Poland.’ Start off by thinking, ‘What are the needs of the people?’” Leaders need to address the basics: education, health care, the economy, the health and strength of their democracies. “Work with people who are fighting for values that you believe in,” he said. “And when you organize at the grassroots level around those issues, the rest ... will take place organically.”

Share

“I just wanna watch some berndog”

As engaging as the Q&A was, it had nothing on a second stream of commentary that neither the legislators nor the audience was aware of. On the side of the live stream on YouTube was a running live chat. At first I paid no attention to it, but during the lull between the formal remarks and the student Q&A, I scrolled through it. There were a few legitimate comments referencing the actual discussion, but most seemed to be people (and/or bots) who were apparently betting on the number of times specific words were said. “How many times did he say billionaire???” said one. “Don’t say billioners [sic]” said another, apparently trying to short the word billionaire. “I could buy everything u [sic] own with what Bernie just made me,” bragged another—and in another comment referred to Sanders as “berndog.” I wonder if that was one of the gambling terms or just a spur-of-the-moment riff.

You really need to see it to believe it:

So of course I had to look this up. People are seriously betting on specific words political figures are saying at a student town hall? Yeah, it’s a thing. They’re called “mention markets” where you buy “yes” or “no” shares on individual words or phrases spoken at conferences, talks, and other events. Two such markets are Kalshi and Polymarket, which sound like a breakfast cereal and a bazaar for second-hand polyester clothes. But they’re making some people some very lucrative trades.

I looked back through the chat. “Anyone here from Kalshi?” one user asked; a few others referenced it as well.

So there you have it. If this doesn’t prove to you that the Empire is in decline, I don’t know what does. I gotta say, though, I really love the “berndog.”

One more thing.

At the end of the event, the organizers gifted a pair of Stanford-branded mittens to Senator Sanders, as a nod to the viral photo of Sanders wearing thick mittens on the day of Joe Biden’s inauguration in 2021. I managed to snag a screenshot, but the camera wasn’t zoomed in enough to see the mittens close-up. Well, here it is. The audience loved it :)

Gratitude, always

Sincere thanks to everyone who has supported my work, and warmest of welcomes to all the new readers! Whether you’re a paying subscriber or have bought me a coffee or two, know that your support is felt and appreciated.

[direct link here if the giphy thingie doesn’t work]

Have you heard the latest? A new R&B singer has broken onto the scene and signed a reported $3M deal with a record label. Her name is Xania Monet and she’s a gorgeous Black female singer. 106k subscribers and 1 million plus views on YouTube, 121k followers on Instagram, 357.3k followers on TikTok. Sounds like she’s crushing it.

More exciting news: a new emerging talent in acting is said to be catching a lot of headlines: Tilly Norwood, turning heads among Hollywood agents. Not quite as popular on socials yet, but maybe that’s because she’s still getting her headshots done.

Just one little detail: Xania and Tilly are both one hundred percent AI-generated avatars, voice and image. Not a real bone or living cell in their body.

Strange world we’re trying to live in these days. Perhaps the media can help make some sense of this. Let’s take a look at some of the headlines as they’ve come in:

“AI Artist Xania Monet’s Multi-Million Dollar Record Deal Creates Controversy” (Ebony Magazine, September 22, 2025)

The headline blurs the line between actual people and a collective of moving pixels that resemble a human female singer. First, “Xania Monet” is no more an “artist” than the engine in your car, for the simple reason that it is an AI avatar that has neither sentience nor consciousness; it1 doesn’t breathe, sleep, or eat; it is not made of flesh and blood. Secondly, the term “AI artist” has been used to refer to humans who use generative AI to produce images and designs, not the digital representation of an artist. If we cannot depend on language to code concepts accurately, what are we talking about? Finally, the record deal isn’t Xania’s. The deal was signed by the person who created her, a poet and songwriter in Mississippi, who, it should be said, didn’t get there alone. She had the backing of Timbaland, a rapper and producer enamored with AI, and used the AI music platform Suno (currently mired in copyright lawsuits) to create Xania’s voice and music.

“Al Singer Xania Monet Just Charted On Billboard, Signed $3M Deal. Is This The Future Of Music?” (Forbes, September 27, 2025)

Same observation applies… “Xania Monet” is not a singer, AI or not. Again, “she” did not sign the record deal. And if you look closely, you’ll see the acronym “AI” in the headline is actually A1, as in the letter A and the number One:

The public is, understandably, monumentally confused. Some comments (and followers) no doubt come from bots, and some people seem to love the novelty, but many are confused or downright sickened by the idea of a non human, AI generated “singer” marketed to them on the same channels as the human artists.

And speaking of real live human musicians, many are running white-hot mad. As they should be. Breaking into the music business is tougher than ever, and competing against a programmed, animated character that can perform literally 24x7 and has zero real-world concerns (things like breathing, eating, drinking, and sleeping, or paying rent and taking care of the kids, on top of going on tour and keeping your voice strong and warmed up), is so far beyond enfuriating we’ll have to invent brand new vocabulary.

Five-time Grammy-nominated singer and songwriter Kehlani sums it up well:

IT, not she, is taking all of the data it’s collected on us and what we want, and is tailoring to us. even down to the voice. [sic]2

If you’d like to dig into the structural details, this video does a good job of teasing apart some of the nuance in this alleged record deal.

Synthetic Stars

We knew it would come, this moment. I can now say where I was when I first heard about an AI generated “actor.” It was just this past Saturday night. I was researching names for a character in a new short story I’m writing (yes actually researching, not asking a chatbot), when I came across the news that an AI generated entity called Tilly Norwood (often misspelled as Norward, perhaps intentionally so—”norward” could easily mean “neither backward nor forward,” which is a rather apt commentary on the role of AI in the arts) was released by a new “AI Talent studio” at the Zurich Summit the same day.

It’s all a bit hush hush apparently: the two women interviewed on the panel, one of whom is the founder of this new AI studio, “suggested that studios and other media and entertainment companies were quietly embracing AI under the radar.”3

I’m not mentioning the names of the creators of these AI avatars because they’ve got enough press; and that’s precisely the point, in both cases. All the press and headlines you see about Xania and Tilly are, in many ways, engineered. The people behind these avatars and these record labels and production companies know full well how polarizing the topic of generative AI in the arts is, how the news and social media algorithms work, and how emotional reactions, whether positive or negative, drive virality. Virality is the point.

Like Xania, Tilly has its own website as well as Instagram, TikTok, Facebook, and— wait for it—LinkedIn user accounts. Little to no indication that this isn’t a human being; the Instagram bio says, haughtily: “You’ll either get it or pretend you don’t. I’m a creation.”

Rather condescending wouldn’t you say? Silly Tilly.4 And bizarrely vague: “I’m a creation”? I’ve got news for you, Ms. Synthetic Entity—I’m a creation too, of my parents, with a fully functioning body and mind. It would be super nice if your creators could be just a tad more up front about the fact that you are not a real human and stop trying to blur the line. We, the human public, aren’t stupid, and those of us who are real actors, even less so—and much more ticked off.

But let’s back up a little lest the usual “Luddite” war cry come down on our heads. CGI or digitally animated (read: non human) characters are nothing new. We’ve laughed with them, we’ve cried with them and for them, we’ve bought the merch and stood in line to see the movies.

But gen AI avatars are not the same thing as CGI characters. They break down the trust we have with stories, because it’s getting a little difficult to tell the difference between a real live human being and an AI generated avatar. (Granted, Xania is much more obviously AI-generated than Tilly, but we’re splitting digital hairs at this point.) These avatars are now being marketed as synthetic humans… injected like a poison dream into our consciousness, telling us, on their social media accounts, that they “had such a blast filming some screen tests recently,” that they “can’t believe it… my first ever role is live.”

But these AI entities don’t do this on their own. They’re not sentient. They do not breathe or think. All this is done by the humans behind them. “Real emotions.” No LLM or algorithm has any conception of what an emotion is, because to feel an emotion, you need a physical living body. So now in addition to fake news, we have fake expression. Fake communication. Fake emotions. Microplastic cocktail, anyone?

Moana never tried to slide into our DMs to flutter her eyelashes at us and invite us to her red carpet premiere. Jake Sully of (the now ironically named AVATAR) isn’t doing side gigs as a TV host or selling us blue ketchup. Moana, Jake Sully, and so many other characters in our favorite CGI movies, live in *their* stories, not our world, and that is a sacred line the people behind the avatars are trying to cross.

You wonder whether they’ve really exhausted their artistic soul to such a point where they can think of nothing more original than to produce an AI character. That soul should run a lot deeper than that. But we’ve all seen what the promise of money and status can do to willing humans.

Like I said, the public isn’t stupid. When you read through the stacks of comments on these AI avatar’s socials, you’ll find plenty of people who are aghast, disgusted, and opposed to the idea of AI music or film characters. So far, the one thing that keeps the avatars on their own deserted digital island is the fact that they cannot hold a scene the way human actors do. Their movements are still plastic; their eye contact is hollow; that connection that you feel with a live performer is missing. In this YouTube short from the making of Avatar 2, with a split-screen showing the human actor above and the final composite scene below, it’s clear where the emotional core of the story on the screen comes from. The human actor. (And the fans agree.)

The danger down the road lies in generations of young people growing up in an ocean of AI generated content, never coming into contact with the real, raw artistry that comes from the lived pain and experience so many time-honored musicians and artists bring to their craft.

This is why it’s so critical to maintain your critical eye, your appreciation of mastery and craft, and your own practice, be it in words, painting, music, photography, or any other artistic effort—and to pass that passion on to your children.

Let’s heed the words of a commenter on one of Xania’s TikTok videos:

“We in trouble y’all.”

I have neither a $3M record label deal (given my utter inability to sing, that’s actually a good thing) nor an AI avatar writing for me, so any form of appreciation of the work I do here at The Muse is met with gratitude. You can become a paying subscriber or buy me a coffee so I have enough energy for the next essay. :)

[direct link here if the giphy thingie doesn’t work!]

On May 6, over 25 creatives woke up, had breakfast, got in their planes, trains, and automobiles, and headed for the State Capitol in Sacramento. Made their way through security, down an ornate, impressively quiet hallway, into a well appointed vintage elevator, and up a few floors into a small room.

There they sat, patiently waiting.

Writers. Artists. Musicians. Singers. Voice actors. Art directors. Designers. Concept artists. Illustrators. Comic artists. Animators. Art educators.

The Assembly Judiciary Committee would begin soon, and the bill the creatives had come to support was first on the agenda. They were excited, a little nervous, but determined. Some had prepared short statements to read. Others hadn’t, but simply being in the room was enough. It was bodies on the line.

But when the hearing began, another bill was called. We assumed our bill was still getting its makeup done (those of us in the movie biz know that reality all too well). All good, we thought, we’ll surely be next.

But we weren’t. A different bill was called after that. So we sat through that. Then, another bill, and another. There were bills regulating surveillance pricing—the practice of altering the prices of products on the shelves in real time, based on a customer’s cellphone location and their profile. A bill addressing sex discrimination in our post secondary schools. A bill addressing involuntary commitment for people deemed “a danger to themselves or society.” A bill on the language requirements for residential rental properties. And on and on.

Somehow, the original order of the bills for this hearing had been changed. But we knew that might happen—it sometimes does. And so you wait. But in waiting, and sitting quietly, you also get an insight into all the other bills that the committee works through. You see what other people in your community—your neighborhood, your town, your state—care about. What’s hurting or troubling them, and how they think it can be fixed. What’s important to them. You hear the arguments for and against the proposed legislation. You hear the Assemblymembers speak, their expert witnesses, and also the members of the public.

”This is how law becomes law,” I told my high school-age daughter, who was with me in the room that day. “You’re at the heart of it right here.”

Nearly two hours later, our bill, AB 412, was called up. You could feel a perceptible wave pulse through the room, as all of us perked up. It was time.

Tim Friedlander, the head of the National Association of Voice Actors, stole the show with his passionate call to protect the rights of creatives, and a voice made of honey emeralds that mesmerized the committee members (this wasn’t Tim’s first rodeo). Another expert witness, an IP attorney, also spoke on behalf of the bill.

In essence, AB 412 requires developers of generative artificial intelligence models to document the sources of the data they use to train those models, and make that information available to rights holders so they can make informed decisions about how their intellectual property should or may be used. What we creatives call “art” or “our work,” developers call “data.” Paintings reduced to pixels. Novels reduced to tokens. Music reduced to bits of sound.

Simply put, the bill calls for transparency. Not surprisingly, certain arms of the tech industry have been fighting this bill tooth, nail, and limb. Gorging on the work of creative professionals for free, with no accountability, is simply too tempting.

After the opening arguments, the Committee called on the public.

"Is anyone else here in support of AB 412?" asked the Chair.

We all stood up at once. Easily 70% of the room. An audible gasp rushed through the audience as we stood, and approached the mic one by one. We gave our names. Our titles. From the very senior (a retired comic illustrator whose career started with DC Comics in the 1970's) to the very young (my own teen artist and musician daughter), we spoke up, loud and clear. We expressed our undying, firm support for this bill to safeguard our rights as creatives, whether amateur or professional, in the face of Big AI’s relentless efforts to twist the Fair Use arm of Copyright and be granted carte blanche to scrape our work with impunity and zero accountability.

I’m happy to share AB 412 received the votes it needed to advance to the next stage.

But with Shira Perlmutter, Register of Copyrights and Director of the U.S. Copyright Office having been unexpectedly let go this weekend—not long after her office released a preview report that did not cozy up to Big AI, and just days after Dr. Carla Hayden, the Librarian of Congress, was also released from her duties—the stakes of this battle have only increased.

Hold this reality in your heart. The administration fired our Librarian of Congress, and the Director of the U.S. Copyright Office.

In addition to that, a bit of text was slipped into the Budget Reconciliation bill by House Republicans late on Sunday night. Yes, on the night of Mother’s Day:

“...no State or political subdivision thereof may enforce any law or regulation regulating artificial intelligence models, artificial intelligence systems, or automated decision systems during the 10 year period beginning on the date of the enactment of this Act.” (Source: 404 Media)

So… they want to prohibit states from regulating artificial intelligence systems in any way. For a decade.

So much for states’ rights.

Some might ask, why bother. All those tired tropes of “AI is here to stay, why fight it?” We’re all such dinosaurs for not seeing the brilliant lights of a technofeudal future dreamed up by billionaire executives.

To draw inspiration from the words of another nation’s hero, I’d rather die standing up for my fellow creatives than relinquish my life’s work and passion to an algorithm. Much more than hope, we must hold our convictions firm. The right to create, the right to control your work, and the right to earn your living from your own work, are inalienable moral rights. This runs deeper than legislation, farther than any economic measure. We’re standing up for our very right to breathe. 🙏

If you’d like to watch the proceedings, here’s the link. The proceedings for AB 412 start at the 02:19:12 mark.

And if you’d like to join us for the next session or submit a letter of support, DM me.

Forgive the somewhat dramatic subhead. But considering the very real implications of a new legal framework governing copyright and generative AI, if we let our guard down, the authorities & the AI corporations will twist the knife in our back, with an extra dose of Maldon salt.

Let’s get right into it, shall we?

The UK government has opened a public consultation for its legal framework governing generative AI and copyright that “supports both the AI sector and the creative industries.”1 The consultation is published by the UK’s Intellectual Property Office (also known by the rather ironic acronym “IPO”), the Department for Science, Innovation and Technology and the Department for Culture, Media and Sport.

First, a quick overview of what is at stake, the deadline to submit a comment, and a few things to keep in mind before you do.

tl;dr

At issue: The legal and practical definition of copyright, and the ability of AI companies to use your work without consent, credit, or compensation, and build technology or systems with the potential to compete with you in the marketplace

Link: The open consultation is here.

Deadline for comment: February 25, 2025, 11:59 pm UK time

How to comment: Via email or the online link

Who can comment: Anyone who is a citizen or resident of the United Kingdom. And also, apparently, anyone else outside the UK who cares about this topic—nothing I have seen on the proposal website or elsewhere explicitly states that only UK citizens or residents may respond. For additional due diligence, I contacted a UK IP attorney (last name, Mr. Bond) and he said he wasn’t aware of any restrictions in this regard. So please also share and forward widely to all your friends and colleagues in and outside the UK.

Why to comment: We thought copyright would always be there to protect our work. That may no longer be the case, in practice, if we allow this type of legislation to go through.

Read the entire post. Please take the time. Resist social media, wailing children, the scent of Christmas cookies, and other distractions for this. I cannot emphasize enough how critical this is.

Before you comment: If you’ve commented on government legislation before, you’re likely familiar with how it all works, and what types of comments tend to be more useful than others. If this is your first time participating in a public comment or consultation, all the more important to read at minimum this post before you submit your comments.

**Above all, avoid making the argument against the usage of generative AI in general. This is not what this legislation is about, and such comments are likely to be ignored. As UK-based animator Jodie Hudson explains [emphasis mine], “The question the UK wants feedback on is not whether AI will be used in the creative industries or not, but how it’s used in the creative industries.”

Lastly, please do bother making a comment if you are able. This one act you do is potentially more meaningful and impactful than any online debates or discussions you can engage in, because this legislation will hold legal, material weight2; online debates do not.

🌰 In a bigger nutshell

What is copyright?

We use the term on a regular basis, but do we really know what it means, especially in its legal sense? Disclaimer needed here: I’m not an attorney, so please feel free to consult one, but here is a definition from the U.S. Copyright Office [emphasis and the exclamation point theirs]:

Copyright is a type of intellectual property that protects original works of authorship as soon as an author fixes the work in a tangible form of expression. In copyright law, there are a lot of different types of works, including paintings, photographs, illustrations, musical compositions, sound recordings, computer programs, books, poems, blog posts, movies, architectural works, plays, and so much more!

In other words, that idea you have for a blockbuster novel, a breakout new song, or a blueprint for your city’s new opera hall? Not copyrightable until you render it in some tangible format. So if you tell someone about your idea, and they flesh it out on paper, in video, or another tangible form (this includes digital) and copyright it, you’re out of luck. (I doubt an NDA would fully protect you either, but again, consult legal counsel.)

As for Fair Use, the goblin that’s been a thorn in the side of many a discussion and debate—and a perfumed plastic rose for generative AI models—it’s considerably more complicated. Specifically, it is Section 107 of the Copyright Act that outlines the scope of Fair Use; it calls for the consideration of the following four factors:

1. The purpose and character of the usage—is the intended use of a commercial, nonprofit, or educational nature?

2. The nature of the copyrighted work—to what degree does the work encourage creative expression? Is it highly artistic or more factual?

3. The amount and substance of the portion of the copyrighted work used—how much of the work, and which portion or element of the work were used?

4. The effect on the potential market for or value of the copyrighted work—this speaks for itself, but essentially, does unlicensed use harm the market or economic value of the copyrighted work, whether present or future?

In brief, what a rose is to copyright, a bramble bush is to fair use. Here’s an official rabbit hole for your reading pleasure.

Note that legally, copyright is distinct from trademarks and patents.

Who should be concerned

Copyright matters to anyone who creates anything in tangible (material or digital) form for a living or as a hobby, for example people who:

• Write poetry, essays, books, articles, and other fiction or non fiction

• Draw, paint, illustrate, design, photograph, or create any form of art in 2D or 3D

• Compose (and perform) original songs or musical works

• Write, direct, or produce films, documentaries, or other audiovisual media

• Create code for computer or software programs

• Obtain ownership rights through “work for hire” contracts

In the eyes of the U.S. Copyright Office, anyone who creates an original work and fixes it in a tangible form, is a copyright owner. Critically, “Copyright exists automatically in an original work of authorship once it is fixed.”

Registering your work with the Copyright Office gives you specific additional rights, such as the ability to enforce your exclusive rights of copyright via litigation and to facilitate licensing fees for your work by enabling others to find the rightful copyright owner of a work they wish to use. Notably, it also does something that’s rarely brought up in these discussions:

Copyright registration … provides a record of this nation’s creativity.

So, creativity patriots of America, register your work! Make America Creative Again!

Meanwhile… across the pond

The UK government has recently opened up a consultation period for a new round of updates to its copyright statutes. It seeks comments from the public on critical issues such as digital replicas, “copyright protection for computer-generated works” and the following objectives:

Boosting trust and transparency between sectors, by ensuring AI developers provide right holders with greater clarity about how they use their material

Enhancing right holders’ control over whether or not their works are used to train AI models, and their ability to be paid for its use where they so wish

Ensuring AI developers have access to high-quality material to train leading AI models in the UK and support innovation across the UK AI sector

It sounds good, on first impression, doesn’t it dear. Very diplomatically worded. You can almost smell the Earl Grey and the croissants. As with all legal proposals that say anything about “ensuring AI developers have access to high-quality material” and “supporting innovation in the AI sector,” one might indeed wish to brush aside the pleasantries and take a long hard look at the implications buried within. As an example, in Section B.2 AI training and copyright, Article 42 states:

“…in many cases, AI models are trained using works made available to the public on the Internet. These are often not expressly licensed for AI model training, and the creators of those works are not compensated for their use.”

Article 45:

“Some AI developers argue that existing legal exceptions in UK copyright law allow them to use copyright works when conducting training activity in the UK. If, their training activities take place in other countries, they may argue they are not subject to UK jurisdiction. But right holders reject these arguments. They maintain that, by making copies of their works to train models, AI developers are infringing their copyright in the UK.”

Article 47 assures us that the intention of the proposed legislation is to “ensure that both the AI and creative industries can share in the benefits of AI, and that both sectors are able to grow together.”

That’s really great. Maybe we can all play nicely together in the sandbox. But then things take a somewhat darker turn. Article 48:

The government does not believe that waiting for ongoing legal cases to resolve will provide the certainty that our AI and creative industries need in a timely fashion, or, potentially, at all. We are therefore considering a more direct intervention through legislation to clarify the rules in this area and establish a fair balance in law.

The government “does not believe” in waiting for lawsuits to resolve. Big Tech is pretty impatient too. They want their high-quality data, and they want it now. What better way to get out in front of all those pesky, snail-paced legal proceedings than to effect a “more direct intervention”? This sounds an awful lot like changing the law ahead of the lawsuit.

The proposal presents four options, bizarrely presented as Options 0, 1, 2, and 3. I can’t help but take “Option Zero” to imply that we should simply skip over this one. This is how they’re introduced:

Option 0: Do nothing: Copyright and related laws remain as they are.

Option 1: Strengthen copyright requiring licensing in all cases

Option 2: A broad data mining exception

Option 3: A data mining exception which allows right holders to reserve their rights, underpinned by supporting measures on transparency

Option 1 offers the strongest protection for rights holders (and cries about being less competitive in the AI sector as a result). Option 2 is the copyright killer—it would allow “data mining on copyright works – including for AI training – without right holders’ permission.” This would effefctively turn “fair use” into “unfair use.” Option 3 vaguely refers to a position of compromise, by having rights holders essentially have to opt their work out of AI training. Which, as we all know based on prior experience (cough Meta cough OpenAI cough so many others), isn’t going to work. AI developers have already scraped billions of words, images, music data, and minutes of video without permission or authorization—the onus should not be on us creatives to ask to “reserve” or “request the protection of” our rights. Copyright has historically been granted automatically, not as the result of an obligatory act of reserving your rights. As Article 67 of Option 3 states [emphasis mine]:

“AI developers would be able to train on material to which they have lawful access, but only to the extent that right holders had not expressly reserved their rights.”

I urge you to read the full texts of these options for yourself—they’re not long, and they are eye-opening. Also see Why you need to pay attention below.

Comments are welcome from creators, rights holders, and AI developers to help shape a copyright framework that allows both sectors to thrive. Remember the deadline is February 25, 2025. That date approaches furiously. We fully expect Big Tech to flood the comment form with arguments to let them bulldoze over creatives’ work with abandon, hence the next segment:

Why you need to pay attention

Don’t live in the UK? Right, so who cares what happens across the pond! You don’t drink tea anyway. Well. I doubt I have to explain how critical it is that we here in the U.S. keep close tabs on copyright legislation taking shape in other countries, especially countries as close to us as the UK. Because…

Once copyright protection is taken away or diluted, you can kiss your creative career good-bye.

Forbes contributor Virginie Berger explains:

The [UK] proposal includes a broad new copyright exception for commercial generative AI training, allowing AI companies to train on British copyrighted works without a license. Central to the controversy is the introduction of an opt-out system for copyright holders, which would permit AI companies to use copyrighted material by default unless rights holders explicitly request to opt out.

This polarizing move has sparked fierce criticism from creative industries, who argue that the proposed changes favor big tech at the expense of protecting creators and intellectual property.

We already know how “opt-out” works. It doesn’t. You create, produce, and publish; they scrape, train, and monetize. Whether or not it’s legal or ethical. Don’t like that? You’ll have to send in your ample legal resources, which costs money and time and… yeah.

But it’s not just your creative career. With generative AI and the speed with which it’s evolving, our regulatory troubles won’t end there. There’s a concept called digital replica, defined generally as a reproduction of copy of your voice and/or physical likeness created with digital technology, such as artificial intelligence systems.

SAG-AFTRA, NAVA, and other groups representing professional actors are now fighting ferociously to protect their rights to their digital replicas. As well they should. You can’t just go out and buy a new face, you know? Botox aside, you’re stuck with the one ya got, pretty much for life. It’s a battle that could, all too easily and all too swiftly, come to have real-world implications for the rest of us—anyone who happens to have a face, a voice, and a body. Stay tuned on that. For now, be very, painfully, acutely aware of the slowly tilting, oil-slicked slope of the copyright debate. Because loss of copyright protection for your works will soon morph into loss of protection for your very identity. Let’s make sure the pull of profits—er, gravity and the oil don’t overwhelm our ability to stand on our own two feet.

A basic human right

It can be argued that there is a higher, moral stance to be taken vis-à-vis ownership of a work. Legal structures aside, most of us creatives agree that the works we produce are, and should be, ours, by nature and by default. If I write a book, it is my book. If I paint a painting, it is my painting. If I compose a song, that is my song. I should have a say where my work is displayed, distributed, or performed, whether it is altered, how much I want to charge for it, and certainly whether someone else can claim authorship to it.

Creative ownership is, and should be, an inalienable human right. Perhaps you disagree with this, and take the stance that this would be true only for those who subjectively subscribe to this statement. Now, if you personally take the stance that creative ownership is not an inherent right, and you’re willing to give up your ownership rights to others without your authorization, permission or knowledge, that is on you and you only. You cannot impose this stance on others who disagree with you, much as you cannot impose your beliefs of any other kind (bodily autonomy, for example) on others. That is, typically, considered dictating, controlling, or oppressing others. Gosh where have we seen that before… besides throughout all of human history…

I digress.

Bottom line, it’s up to us, yet again, to defend and protect our rights as creators, no matter how rationally inalienable and fundamental they may be.

What it’s not about

The proposed UK legislation is not an argument about whether to use generative AI or not. Schrödinger already dropped the bag, and The Cat’s out. For those wondering whether it’s alive or dead, it’s both and neither: it’s a zombie, and it’s gone feral. Lock your doors at night people.

Besides, we cannot stuff the LLM genies back into their bottles. Our—ok, some people’s—three wishes have been granted:

• Produce “content” at lightning speed and ludicrous volumes

• Claim respected titles like “artist,” “musician,” and “author” without having to put in the years of experience, thought, and hard work to deserve it

• Triple our earning power while decimating the livelihoods of our fellow creative humans (or at least give us the next new shiny thing to debate ad nauseum online)

This regulation that’s on the table is more critical than the debate of whether to AI or not to AI. The protections afforded by the very essence of copyright are on the chopping block. If we allow AI companies to train on our work, whether or not it’s copyrighted, the protections of ownership automatically granted to every creator, whether or not they formally register their work, will essentially be nullified. And then the only way to protect your work will be to write it out on paper with squid ink and hand-bind it, or have it printed without ever touching the Internet. Is that the future we want?

Rabbit holes to explore

The text of the UK Government’s Copyright, Designs and Patents Act 1988 is here. It’s long. Really long. I wouldn’t wish it upon my worst enemy. Suggested additional reading: author Kate Mosse’s insightful article in the Financial Times about AI and copyright (thanks to Ed Newton-Rex for the suggestion).

I also recommend Virginie Berger’s article in Forbes, quoted above: “What the U.K.’s AI Copyright Reform Means for 2025 and Beyond.” Virginie is a Los Angeles-based music & tech business professional and writer who covers AI and rights management.

For additional insights and template examples of responses you can customize, see Ed Newton-Rex’s article on the topic here on Substack and his posts on BlueSky. (For more context, Ed was Head of Audio at Stability AI before he left due to his personal opposition to the company’s “exploitative” practices of using creatives’ works to train their models without consent. Ed is also behind the simply titled but massively supported “Statement on AI Training”, which I heartily encourage creatives to sign.)

The (tragically) late Suchir Balaji’s post “When does generative AI qualify for fair use?” which goes into some illuminating detail about how generative models are trained on “copies of copyrighted data.” Suchir, you might have heard, was a brilliant young engineer who played a key role in the development of OpenAI’s ChatGPT and later turned whistleblower against the company on ethical grounds. He died this November 26 of an apparent suicide. This was just after he had been named in The New York Times lawsuit against OpenAI as someone who had important documents for the newspaper’s case. Needless to say, theories are already swirling…

If you have additional resources you’d like me to add, please DM me or put it into the comments below.

One more thing.

I know it’s the holidays. Many of you are traveling, running around doing last-minute shopping, preparing dinner menus. It’s more than a little inconvenient that this “open consultation” period was sprung upon us on December 17, a mere seven wondrous and blessed days before Christmas Eve. If you were Stephen King, perhaps you would think it was some nefarious plot cooked up by the UK politicians to ensure we have as little time as possible to comment. But that’s just rubbish isn’t it… I’m sure they were just really busy all year. Nevertheless, do make sure you don your White Knight of Copyright regalia, and ride into the New Year with a thoughtful comment or two for the Lords.

To hear some people tell it, generative AI is the greatest thing since... the invention of the paintbrush?  

Perhaps, or perhaps not. Plenty of ink and ire have been spilled over the dubious practices of Big AI—the tech firms that have vacuumed up images, text, music, and video from all over the Internet to feed and train their generative AI applications. This has resulted in massive heated debates, worker strikes, open letters signed by thousands of creatives, and more than a few lawsuits. That’s what has been ruling the headlines for the past two years.

As a result, we now have lots of stats on the number of IP & copyright lawsuits, the sizes of the training data sets, and the numbers of creatives who’ve jumped ship to Cara and similar platforms.

What’s a little harder to gauge is the personal, individual impacts of generative AI on creatives—especially our ability to practice our craft and earn a living from it. Is it a revolutionary tool for creatives or is it a massively overhyped technology that’s best deployed in other ways? We’re also not hearing—on any meaningful level—the stories of creatives who are personally and financially impacted by the technology.

A group of us have decided to ask these more nuanced questions, and to tell these personal stories. The effort is two-part: a questionnaire and a series of 1:1 interviews, culminating in a narrative report.

How it’s set up:

Part I —  Basic information about you and the kind of creative work you do.

Part II — Your experience with generative AI, whether or not you've used it in your work.

Part III – Your take on generative AI's impact on your industry.

Part IV — Assessing the time you've spent addressing gen AI-related issues in your life and work.

Part V — Assessing the impact of gen AI on the value of your talent and work.

To fill out the survey, tap on the button below. If you’d like to be considered for an interview and your story to be included in the report, leave your contact details at the end of the survey.

Go directly to survey

What will you do with the results?

Once the results are gathered, crushed, fermented, and bottled, we will turn them into a fine report and publish it here on The Muse—as a start. The report will also be made available to mainstream and independent media, regulators, policymakers, trade and professional organizations, and other interested parties working to protect the creative classes from harmful impacts of generative AI, whether direct or indirect.

And yes, we’ll also send the report to everyone who takes part in the survey.

How long will this take me?

It looks long and daunting, but each part is fairly short and will take you less time than feeding your pet dragon. In more specific terms, about 10 minutes. We’ve also tried to make it non boring.

Who are you people?

The entity managing this study is LUCITÀ, a small hybrid media publishing and consultancy firm based in Northern California. The report will be written by Birgitte Rasine, founder of LUCITÀ, literary author, and journalist (she used to cover the world of Hollywood and digital camera technologies for The Hollywood Reporter and Daily Variety, and the WGA strikes of 2000 for BusinessWeek); her work has also been published in HuffPost, Global Focus (Europe), Central America Today, and other publications.) Birgitte also writes "The Muse," (this newsletter!) about human ingenuity and creativity in the high-tech era (which has recently been bulldozed over by generative AI).

Special thanks goes to Johan Cedmar-Brandstedt (a fellow Substack author), animator Jodie Hudson, computer science professor Dr. Ben Zhao, entrepreneur Dima Durah, photographer Robert Kneschke, and others.

We are not a major research firm or consulting group. We’re not funded by deep pockets. We’re doing this on our own personal time and with the help of people like you helping to spread the word (gratitude!).

Do you have an agenda?

There is no agenda, political, financial, or otherwise. No one told us to do this survey, and no one is paying us to do it. The original idea sparked from a chat with some of my fellow creatives. I noted how much time we creatives seem to be spending on various aspects of the gen AI debate, and as a result how much time we’re not spending actually creating, honing our craft, or working on projects—and the fact that those kinds of impacts haven’t really been addressed. And because I like to get answers rather than just guess why something is or what things seem to be, we decided to do this survey.

By the way. As is typical in the generative AI debate, our project seems to have fomented the ire of a few people who seem bent on invalidating our efforts, insisting that the survey is “biased” and that it “wouldn’t pass academic certification.” One person has warned that if you take this survey, “you have to be very careful.” Indeed. The survey is one long obstacle course, littered with medieval torture devices, hollering witches, and ticking clocks. It makes you think. Reflect. Worse still, it lets you express your true unbridled sentiments about a technology that apparently all of us are just supposed to swallow hook, line and sinker with zero questions or nuanced discussion.

Re: the argument about “bias,” note that the survey includes positively worded statements about generative AI. Love genAI? Great. Hate it? Hear you. Not sure? Also valid. So, fill it out—truthfully, honestly, with no fear, and on your own recognizance!

Who else is participating?

We can’t share specifics (see the part below about privacy), but we can share that respondents hail from North America, Latin America, Europe, Oceania and Asia, are members of well-known guilds and professional organizations, and include lifelong professionals as well as new entrants to their fields.

Okay, fine. I’ll do it. When’s the deadline?

While at this point we have a robust number of respondents, we're going to keep the survey open while we conduct follow-on interviews. We would also love it if you shared the survey with your artist/creative friends. Because in this party, the more the much, much merrier.

Here’s the survey link you can share. Or hit this button:

Take the survey

Please note! This survey is designed for creative professionals only. It is not intended for executives, producers, marketing & PR people, agents, or others who interact with creatives as part of their jobs. Any PII you submit, such as your name and contact information, will be kept strictly confidential. (We have an army of rabid chihuahuas guarding our hard drives. We’d show you pictures of the damage they can cause to a hacker’s body but it’s simply too gruesome.)

If you'd like to be considered for an individual interview to be included as part of the final report, please be sure to include your name, affiliation, and email address at the end of the questionnaire.

Our warmest appreciation,

Birgitte and team

Mere hours ago, I met with a friend of mine I hadn’t seen in a few years, despite us living well within a short driving distance of each other. If you’re also a mom, you’ll immediately understand—time melts into a chaotic bubble of plasma when you’ve got kids and family and afterschool classes and work and writing and …

Living here in the Bay Area, of course we chatted about, well, the chatbots. The generative AI now permeating seemingly every aspect of life.

Yes, we have family members working in high tech; yes, we have plenty of devices at home, but that doesn’t mean we blindly welcome every new widget into our families’ lives just because a billionaire decided it would be great to make a few more billion with flashy new tech. In fact, the closer you are to the tech (and you don’t get much closer than breathing the rarefied air here in the Valley), the more skeptical you tend to be, especially when it comes to your kids. It’s fairly well known in tech circles that luminaries like Steve Jobs and Bill Gates restricted their children’s access to the very technology their companies produced—in Steve’s case, iPads weren’t allowed in the home at all because, in his words, “We think it’s too dangerous for them in effect.”

Let that sink in: Steve Jobs thought iPads were “too dangerous” for his kids.

Somehow the mic drop heard around Silicon Valley hasn’t ricocheted around the world yet. And that’s because the endless and well-funded waves of hype easily drown out the voices of caution and critical thought. It’s been bad enough with social media; the gen AI craze has been nothing like any of us have ever seen.

And yet, the tide is turning. People are speaking up more. Companies are rethinking their AI strategies. Brands are prohibiting their advertising agencies from using AI. The lawsuits keep coming. Most critically, the public is becoming more and more aware of the bad illegal unethical and life- and career-damaging things you can do with a little deepfake.

A year and a half after the launch of ChatGPT, enough people have grown frustrated enough about the deleterious actual and potential impacts of gen AI that they’re taking action, on multiple fronts. One group that has gotten a lot of traction is PauseAI, a collective of volunteers and local communities coordinated by a non profit based in The Netherlands (yes, an actual non profit, not one with a hidden for-profit motive).

Before you make any fractional-second value judgment of PauseAI, before you dismiss them as one of those stuck-in-a-horse-carriage anti-tech groups, take the time to dip below the headline and take in what they are actually advocating for. Take the time to read their FAQs. You’ll see fairly quickly that this isn’t Elon’s version of “pausing AI so I can go make my own because I need to catch up to blow past my tech bro bro’s.” This is a call to slow down a little, and do something that’s becoming more radical every day: take the time to think through what we’re doing, and how it might impact the lives of our own children.

Subscribe now

One of my readers, who also happens to be a member of PauseAI, wrote a heartfelt piece on Reddit about the anti-democratic nature of genAI. I suggested to him to post it on Substack, and that suggestion soon turned into an offer to publish it here. Please welcome to our online stage Sean Pan, author of the first guest post for The Muse.

by Sean Pan

AI is an obvious threat to human creativity and an existential threat to human existence. There is, unfortunately, nothing particularly sci-fi about it. We only have to realize how it is already trending, how it is already harmful to humanity and how it does not, in any way, do the thing that its boosters falsely claim that it does, which is to say, democratize.

What is actually does right now is to destroy the entire ecosystem that it is in; for example, the artist is perhaps disempowered from being able to empower and create human networks. The prompter, is in fact, not empowered; in practice, their content is essentially wiped out en masse. If you want to make a Luddite analogy, creating more poor quality clothes harmed the status of tailors, weavers and women; it did not, however, enhance the status of the boys forced to work the dangerous mills or even the sweatshop workers to this day, it only really concentrated power among the wealthiest who could afford the capital.

So that is what AI is doing right now, destroying the artistic and creative ecosystems via pollution.

What is the goal for AI even in the medium term? To replace as much human work as possible, which is to say, to disempower humans from any economic need and viability. Often this is portrayed as a problem with capitalism, but I find that to be irrelevant because the people pushing this right now are indeed using capitalism and almost sadistically, suggesting that "we might have utopia" whiel causing us immense harm at the moment. Almost 80% of Americans want AI to be slowed down, but instead, we are seeing the companies literally force it upon us, often with pat comments like “for humanity” while literally buying out via lobbying, spending almost $40 million in lobbying to civil society groups which haven’t quite spent $1 million.

So there’s an entire level of non consensual creepiness to this, where we are forced to accept something and told that this is good for us. This doesn’t even vaguely pass the smell test; it stinks of the basic logic that if the initial attitude of a trend is to treat us badly and make us weaker, the likely future is one of further abuse.

Or perhaps as some of the technology boosters enthuse, not just of abuse, but of extinction. In their positive sci-fi dreams, they suggest that life will be immortal but also that AI is inevitable and you are stupid for even pushing back, so the logical endpoint might be “we all die for AI successors” or the “upload your mind into AI and you can be an undead memory of yourself forever.” Which is to say, to kill us all so that their vision of undead machines ghosting to each other will finally be true.

Whatever.

What is obvious that in their ideal situation is that we are thoroughly disempowered, unable to fight back, unable to add value, and hoping that “they” (the elite? the companies? the government? the machines?) will be nice to us, even though they have demonstrated callousness each step of the way. It’s essentially a sadistic game of starting with destroying value in human lives, in human networks and in human values and ending, apparently, with destroying human lives outright for the better good of machines. But we aren’t disempowered yet as humans. We can still fight back, and if you are among the vast majority of humans reading this who like the idea of living networks, of life continuing to exist and of having humanity continue in the future, join PauseAI like I have. Because to get to a good future, we are not going to get there by racing off the cliff, but only via a fundamentally democratic agreement to move forward in an way that actually benefits us, and not for a fringe of people who are, essentially, extinctionists. PauseAI has protests coming up internationally, including one next Monday, May 13.

Fatalism is not the way to take us to good place with such dangerous technology. We can find a better path forward.

About the author:

Sean Pan is technology enthusiast who has worked in the field for over twenty years who has come to reconsider the value of unbridled progress in light of the human and moral cost, and the risks of disaster in the near future. As a writer and an artist, he has come to realize the effect of this early on and despite strong efforts to research the counterarguments, has found those arguments to be so effective that he is now an active member of PauseAI.

On February 4th, I published an essay titled “AIs Wide Shut.” Like a good spicy dish, it was peppered with film and television Easter eggs. I promised to reveal all eleven of them, and that is the purpose of this post.

But this isn’t a simple reveal-the-eggs-and-be-done-with-it post. A commenter on the original essay did something a little bit unexpected and a whole lot revealing… revealing of just how unreliable the chatbots are when it comes to just about any- helping you take the easy way out in an Easter egg hunt.

The commenter had fed the essay into GPT-4, Gemini Pro, and Claude, to “see if it can identify 11 movies and TV shows.” He went on to say:

Here is the response (fair but struggles, especially when pattern matching fails to find a direct or an excellent indirect match, and you will see humans will do a better job once you get responses identifying movies. AI has some way to go, and I do not think AGI is coming anytime soon. We first need a definition of AGI and then new algorithms and probably a lot more computing). … I used the same prompt but [received] vastly different answers.

Indeed. If you think a pattern matching algorithm can upstage humans who’ve grown up with TV shows and movies, you must be… a chatbot. Apart from that eyebrow I raised at having had my essay fed into three generative AI systems without my permission, I did find it quite curious to see the wide-ranging responses in all three cases, and even more curious that each bot got only TWO Easter eggs correct, out of the eleven. Yes, each bot got precisely two correct—the obvious one and one other (in each case different), despite the fact that several of the Easter eggs were very obvious and drew directly upon character quotes. The full answer key is at the end, but first, let’s have some fun at the chatbots’ expense.

Easter Egg Hunt: ChatGPT answers

(verbatim from the bot as posted by the commenter; notes in brackets are mine)

The document provided [i.e., my essay] does not explicitly mention any movies or TV shows directly by name [no, really, ChatGPT? That’s the whole idea of Easter eggs!]. However, it does contain phrases and references that may be associated with certain movies or TV shows, known for their themes, dialogues, or cultural impact.

Here are the references that might be indirectly associated with movies or TV shows:

1. **“AIs Wide Shut”** – This title seems to play on the title of the film “Eyes Wide Shut” directed by Stanley Kubrick. [Congrats. You guessed the give-away.]

2. **“And quite frankly, my dear, they don’t give a damn.”** – This is a famous line from the movie “Gone with the Wind.” [Yes, it is, and now you’ve given up one of your training data packets! I’m calling MGM.]

3. **“No reason to insist we continue to use manual typewriters instead of word processing programs”** – This could reference the historical context of movies like "His Girl Friday" or "All the President's Men" where typewriters are prominently used. [BR: It could, but it doesn’t, primarily because a generic reference to manual typewriters is fantastically boring.]

4. **“Why walk across the room when you can simply exercise your vocal cords?”** – This could indirectly hint at voice-controlled technology often seen in sci-fi movies or shows, such as “Star Trek” with its voice-activated computer systems. [Umm, no.]

5. **“The dinner party turns into a cult orgy”** – The phrase “cult orgy” might be an indirect reference to the film “Eyes Wide Shut” which features a notorious masked ball scene. [Correct reference, good one! But you don’t get a point here because you’re supposed to identify 11 different films or movies. Also, what does it say about a movie when a phrase like “cult orgy” becomes so inextricably intertwined with it that even a chatbot recognizes it? Let’s move on.]

6. **“Spend three weeks on a ship when we can fly”** – This could be a generic reference to the historical context of travel before modern aviation, which is a theme in many films like “Titanic,” although it's not a direct mention. [Let me give you a hint, ChatGPT… to be an film or TV Easter egg, the reference has to embody a quote, a scene, or some other aspect of the show that has become iconic or memorable in some way; or it has to be a hidden message referencing a part of the film relevant to the message. It’s not a theme or concept. No one said anything about three weeks on the Titanic in reference to flying because, well, commercial airplane travel hadn’t quite become a thing yet.]

7. **“Our children whose careers haven’t yet been conceived, never mind prepared for.”** – This could allude to the educational and futuristic themes present in movies like “Dead Poets Society” (in terms of education) or “Minority Report” (in terms of preparing for future careers). [Ooh this is a reeeeal stretch. What did I just say about Easter eggs and themes?]

8. **“…conscious, if AGI will kill us or keep us”** – This theme of artificial general intelligence (AGI) is prominent in movies like “The Terminator” and “Ex Machina.” [Someone is not listening. Sigh.]

9. **“…which artist’s style we’d like to mimic today”** – This could indirectly reference the plot of the show “Westworld,” where the creation of environments and characters based on different themes is a central element. [It could, but it doesn’t, for the reason given in #6.]

10. **“The dark irony is that it’s the people with the least knowledge or awareness of AI”** – This might be an indirect nod to the theme of the knowledge gap between different societal groups portrayed in dystopian movies or series like “The Matrix” or “Black Mirror.” [Forgive us Keanu!]

11. **“Shakespeare would have had a field day with the level of drama now playing out”** – While not a movie or TV show, this is a cultural reference to the drama comparable to Shakespearean plays, which have been adapted into countless films and series. [Alright, well, you tried. A for effort.]

These are inferred references, as the document does not directly name any movies or TV shows. [Thanks for reminding us you have no clue about the concept of Easter eggs, Chat. Appreciate the cute touch with the double asterisks though! You get extra points for effort.]

ChatGPT score: 2 out of 11.

ChatGPT started out pretty well, immediately identifying Eyes Wide Shut as the first Easter egg, and also getting Gone with the Wind correct. But it quickly devolved into simply matching various sentences and phrases in my original essay to the themes of movies (probably based on all the movie and TV reviews and descriptions it hoovered up in its training model), rather than demonstrating an actual understanding of Easter eggs. Which, of course, is in itself understandable: algorithms are not human minds.

Easter Egg Hunt: Gemini Pro answers

(verbatim from the bot as posted by the commenter; notes in brackets are mine)

[Gemini Pro seems to be considerably less loquacious in its responses. Perhaps it was too busy hallucinating content for other users at the time.]

1. Eyes Wide Shut

2. The Godfather

3. The Office

4. The Matrix

5. 2001: A Space Odyssey

6. The Wizard of Oz

7. The Shawshank Redemption

8. Blade Runner

9. The Godfather Part II

10. The Godfather Part III

11. You [I’m sorry, what? Looks like a pattern hiccup to me]

Gemini Pro score: 2 out of 11.

The two Easter eggs that Gemini Pro got correct are Eyes Wide Shut, of course, and The Godfather. How disappointing there was no commentary to respond to! Dammit I should have put in an Easter egg for Blade Runner, one of my favorite films of all time.

Easter Egg Hunt: Claude answers

(verbatim from the bot as posted by the commenter; notes in brackets are mine)

[Claude does provide some commentary to explain how it arrived at its guesses, but ChatGPT still wins in terms of loquaciousness!]

Here are 11 movies or TV shows I identified as being referenced in the attached text:

1. The Shining – References “the password for the House”

2. 2001: A Space Odyssey – References needing a password to enter the house [Here, no doubt Claude mismatched the Eyes Wide Shut reference to 2001, another Kubrick film.]

3. Eyes Wide Shut – Referenced in the title “AIs Wide Shut” [How smart you are Claude]

4. Citizen Kane – References “Rosebud” as the password for the house [A flat-out, straight-up hallucination. “Rosebud” appears precisely ZERO times in my essay.]

5. The Matrix – References being “red pilled” or having one’s eyes opened [Yet another hallucination. No references to red pills in my essay.]

6. Westworld – References AI becoming its own end rather than serving humans

7. A Clockwork Orange – References AI having “incurable hallucinations” [You’re imagining things Claude]

8. Ex Machina – References AI having “irresistible synthetic sex” [Gotta hand it to you, this would have been good except you missed the Easter egg right before and the one right after this phrase. Keep trying!]

9. Succession – References “Succession drama in the House of OpenAI” [Correct! Surprised that neither ChatGPT nor Gemini Pro got this one.]

10. Brave New World – References being “boiled in addicting waters” and “primed” [No frog Easter eggs and certainly no intentional reference to Amazon]

11. Inception – References “don’t unsee AI now” like an idea planted in a dream [No instance of “don’t unsee AI now” in my original essay. This is but a hallucination. How ironic of you Claude.]

Claude score: 2 out of 11.

Claude gets Eyes Wide Shut and Succession correct. I did enjoy reading the commentary ChatGPT and Claude generated, as it gives us a glimpse into what the algorithm is searching for, and how it identifies what it “thinks” are the correct Easter eggs.

Perhaps the reason the scores of all three chatbots are so low is because of the original prompt, which apparently was ““see if [you] can identify 11 movies and TV shows.” Easter eggs are about much more than just identifying a movie or TV show. There’s a lot of cultural context and experience that get baked in as well. Still, given how generative AI works, the bots might have increased their scores a little with different prompts, but I wager we humans would still beat them.

Easter Egg Hunt: the actual answers

1. Eyes Wide Shut (1999), the last film the legendary director Stanley Kubrick ever made. This is (should be) the most obvious Easter egg. The very title of the post is a riff of the film title, and references to it are embedded throughout the post (“eyes wide open,” the password for the House, cult orgies… ). All three chatbots got this Easter egg correct—and they better have. Their LLMs would be really worth less than mud if they hadn’t.

2. Gone with the Wind (1939). An American film classic, directed by Victor Fleming. This Easter egg, embedded in the 5th paragraph (the first full paragraph), is also really obvious—it’s one of the most famous lines from the film, uttered by Clark Gable’s Rhett Butler to Vivien Leigh’s Scarlett O’Hara: “Frankly, my dear, I don’t give a damn.”

3. The Godfather (1972), directed by Francis Ford Coppola (whose hand I once shook at an industry event in Los Angeles). The Godfather series is one of my favorites, an absolute masterpiece of cinema. The Easter egg is in the 6th paragraph: “…one day passwordless authentication might be an offer we won’t be able to refuse.” It refers, of course, to the chilling quote “make him an offer he can’t refuse,” uttered by both Don Vito Corleone (Marlon Brando) and his son Michael (Al Pacino) at different times in the film.

4. There’s something about Mary (1998). This one is a romcom with Cameron Diaz and Ben Stiller that I’ve never seen (and never plan to), but there was something about it that caught my AI 😎. The Easter egg is baked into the first subhead: “There’s something about AI.”

5. Star Wars (1977). On this one, you really need to be paying attention, as it invokes Master Yoda (the last line in the second paragraph below the word cloud graphic): “Being written another essay on that topic is.”

6. Romeo and Juliette (1968). Yes, technically this is a theatre play, but it did get made into a film by Franco Zeffirelli. The Easter egg here is that so often quoted breathless line “Oh Romeo, wherefore art thou Romeo?” — found in subheader number two in the post. And this is, in fact, the richest paragraph of the entire post as far as the Easter eggs are concerned, with five of them embedded in this paragraph (Easter eggs #6-10).

7. A Beautiful Mind (2001). This one just had to be woven in, given what the film is about. If you haven’t seen it, highly recommend. The reference is my line “Five hundred seventy gigabytes of training data produced by the synthesis of countless beautiful (human) minds.”

8. Sixteen Candles (1984). Should be obvious to those of us of a certain age… the hint is in the “…like candles on a young woman’s most embarrassing high school cake” line.

9. Like Water for Chocolate (Como Agua Para Chocolate) (1992). The clue flows through this line: “…like electrified water for chocolate LLM’s.” Bet you thought that was a double Easter egg, the other reference pointing to my other Substack, . You would, of course, be very correct.

10. Succession (2018). This egg lies in the line “Succession drama in the House of OpenAI.” And how very apt now for Stability AI as well, whose CEO has recently been pushed out resigned.

11. A Game of Thrones (2011). Two paragraphs above the “Big Fish, Little Fish” subhead in my essay, you’ll find the line “Never forget that the human mind needs reason like a sword needs a whetstone.” If you’re a Game of Thrones fan, you’ll recognize this Easter egg easily. In Chapter 13 of A Game of Thrones, Tyrion Lannister utters his oft-quote, oft-memed line “A mind needs books like a sword needs a whetstone” in response to Jon Snow who asks him why he reads so much. I found it particularly apt to weave this into an essay about generative AI trained on billions of our words.

And that, my friends, concludes our deep dive into the basket of Easter eggs my henhouse has wrought—and our friends the chatbots made a veritable omelette out of. And, incidentally, blew up what was meant to be a short and simple answer key into a dissertation on the inability of algorithms to intuit film and TV references.

Do you weave Easter eggs into your writing? Or art? Or Excel spreadsheets? Please share! Nothing more fun than hunting for eggs. We could build our own Easter egg house right here on Substack.

Oh one more thing. I’m climbing the beanstalk to get to my golden hen, and I’d love a little push up the stalk:

Donate an Easter egg

In the art world, Ben Zhao needs no introduction. He’s the man running the team behind Glaze and Nightshade, the tools of choice for artists who prefer not to feed their creativity and talent to the maws of the generative AI data scrapers. Officially, he’s the Neubauer Professor of Computer Science and the Director of Graduate Studies in Computer Science at the University of Chicago. Unofficially, he’s a deeply passionate technologist with his heart beating fiercely in the human world.

I spoke to Ben about his work, IP and copyright, the persistent misconceptions about the lives of artists, how generative AI is damaging the way creators earn their living, and the role of friction in the system.

Enjoy! And remember to take your magic potion tonight.

Birgitte Rasine: Ben it’s great to talk to you again and catch up. Congrats on launching Nightshade! First thing I’m curious about, how many people are using it—and Glaze—now?

Ben Zhao: 2.2 million Glaze downloads last I checked, and Nightshade is a quarter million after five days.1 But you know, we don’t keep a running count. I imagine it’s more than that now.

BR: Do you feel there’s anything missing in mainstream media coverage of your work?

BZ: There’s been a lot of media coverage [of our work—see here, here, and here]. I don’t know that there’s anything super obvious that’s been missed but there’s still a prevailing misconception of who artists are and what their daily lives are like. That’s probably the biggest thing that fuels a lot of the opposition to artists, this idea that these are elitist snobs sitting in an art gallery drinking wine and raking in the money. Boy these artists are whiny! Trying to keep the masses from unleashing their creativity.

BR: Yes that’s the big question that’s gotten thrown around a fair bit in the past year—various iterations on the why—why should we protect the work that artists do. To us creators it’s fairly obvious.

BZ: Yeah that is often times the view in the tech sector, but it comes from a lot of misunderstanding of real artistry is and what artists’ lives are like. Over the last 14-15 months, [my team] has interacted with literally thousands of artists. The large majority of artists, even the successful ones, don’t earn that much money. These are all middle class, working class people. Most them I would say are freelancers. They don’t have job security. They work on a contract basis. Some of them live on commissions alone, and so there’s zero sense of stability; any month can be wildly up or down.

So the sense of this entitled elitist group sitting around in art galleries is quite the opposite [of reality]. In fact, even art galleries have a tough time. And when artists deal with art galleries, it’s extremely expensive; art galleries take something like 50% of the profits from the sale. Many artists are already teetering on the edge of being able to support themselves and their families. So disruption like [generative AI] really is an existential threat. How many artists can survive this disruption? You’re not only reducing the amount of commissions people pay individual artists, but also at the corporate level, how many companies are downsizing entire creative departments because budgets are tight and, you know, “we’d like to increase our profit margins. So we’re gonna work with something that looks a little less authentic, a little more glossy and plastic and that’s okay, because profits matter right now.”

So when that happens, the layoffs, the downsizing is just massive. A new report came out recently interviewing different companies, managers and people who control hiring in all these creative fields and they expect something like 260,000 jobs will be disrupted in the next couple of years. It’s just astounding. In California alone, they’re saying 60,000 jobs in entertainment, and then gaming and then illustrations, other sectors, just massive amount of job loss.

BR: Those numbers are huge. But, what they don’t talk about—and should—is the knock-on impacts. When that many people lose their jobs, that affects restaurants, shops, services, and even rentals where they normally spend their money. So, Mr. or Ms CEO, yeah you’ll fatten up your profit margin, but sooner or later it’ll come back and bite you.

BZ: Absolutely. That’s the thing. So much of AI right now is built on hype, and hopes and dreams. There’s the assumption that AI is going to get all fixed in a year, in monumental leaps. They talk about this magical ‘democratization of creativity,’ but nobody’s getting hired. All the companies that are downsizing because they can [do various tasks] with AI, they’re all talking about this increase in productivity, but that’s just a fancy term for ‘your co-workers are gonna get fired, and you’re gonna be doing the work of five people instead of one.’ So all this talk [about democratizing creativity] is questionable and it’s wrecking the entire industry, destroying livelihoods. It’s NOT creating jobs.

I hear this from a fair number of people in these industries who are optimistic and they’re like, ‘I’m gonna be able to work 20% of the time and get paid like I worked 100% of the time.’ Of course, as we all know, this is a capitalist society. You’re not going to reap the benefit and pass it up the chain. Your friends are going to get fired and you’re going to be asked to do entire teams’ worth of work instead of your single job.

And that’s what’s happening. A lot of artists are reporting exactly that—that entire teams are being laid off, and [the people left] are being asked to use AI tools. [The managers are asking], ‘Why can’t you ramp up your productivity by 5x?’ Well, because the AI tools are crap and in many cases, artists have to deconstruct them, then use them to create their own sort of authentic art—and that actually encourages more work, not less. So yeah, there’s just so much misunderstanding and hype.

BR: It’s crazy they’re trying to sell us tools that are not only trained on artists’ work without consent, credit, or compensation, that are not only taking jobs away, but also, insult on top of injury, we’re supposed to pay for them, and they create more managerial work just to get to a result that’s not as good as what the artist would have produced on their own.

So let’s talk about protection. In the cybersecurity business, as you know, there’s the famous cycle of one guy introducing a problem so they can sell a solution for it. If I introduce malware, I’m going to create a market for antivirus software. Now it’s at this fever pitch where hackers and security people are racing each other to the bottom (or top, depending on how you look at it). Do you see anything like this happening in the art sector?

BZ: It’s hard to predict that far down the road because we have nothing in that arena right now. I mean, literally, right now, in IP protection there’s copyright, which is under attack in the public discourse with a lot of people saying, ‘Yeah, copyright’s dead, who needs copyright.’ But if you step away from that discussion, and you look at what’s reality and where the battlefield stands, if you will, there’s nothing on the other side. AI model trainers can train whatever they want. They can download whatever they want. [Now] tools generate prompts from these images and feed them into the pipeline.

What’s on the other side? There are pending legal cases that will take years to play out—and then you’ll have appeals. There are legislation efforts that are hamstrung by companies’ lobbyists and countries’ leadership, not just in the U.S. but across the world, countries that are extremely fearful of losing tax revenue by these supposed booms in AI; so they go soft on AI regulation. Who knows when real legislation will come into the U.S.? Not surprisingly, the EU is moving forward and they’re doing great things. But even that was challenging. Until this week, I think Germany was still on the fence; they could have blocked the AI legislation. Now it looks like they’re going to move forward.

So who knows how long that’s all gonna take? Basically, there is nothing protecting creators right now. There’s nothing stopping companies from ripping off their images and sound and games and characters. In fact, some of that’s happening in real time today.

BR: It’s not a battle if one side has no weapons. It’s a massacre.

BZ: Yeah. Sooner or later people are going to realize this is a one-sided game right now. And all the power is on the side of the model trainers and the AI tech companies. So before we can talk about this balancing act, there has to be something on the other side first, and the only thing that we have so far is Nightshade. Opt-out is the only other option, but it’s not a real option. It’s a great carrot for these model training companies to throw out there. You see them dragged in front of the Senate judiciary hearings on Capitol Hill, and when they get grilled about the ethics of [what they’re doing], they always fall back on opt-out. They say, ‘Oh, we have opt-out lists.’

But how does that work exactly? The artists put their names into a hat, and hope that the companies are going to opt against their own basic for-profit directive, and that out of the goodness of their hearts they’ll respect the wishes of these creatives who have absolutely no leverage. There’s no validation, there’s no enforcement. You can’t check to see if anyone has violated opt-out. Opt-outs are an empty promise, a hope and a dream that a company is going to [sacrifice] their profit margins and say, ‘Yeah, we’re going to intentionally and willingly downgrade our models, and take on the complexity and cost it takes to implement something that filters people out. We’re gonna do all of that just so we can be nice to you even though there’s absolutely no way that anyone can actually verify that we’ve respected or violated opt-out because, you know, it’s a paper list.’

Right now, the technology is not there for us to look at a model and be able to say whether it got trained on a particular image. We just can’t prove it.

BR: Opt-outs are pretty disingenuous, in my humble opinion. There’s a term bubbling up now: AI washing. Like greenwashing but related to artificial intelligence development or usage.

BZ: Yeah and even in the most positive and optimistic terms, opt-out still only involves the companies that have reputational risk—the companies that get dragged in front of the Senate. All the smaller companies, what do they care? They don’t have any risks. They have no need to offer any sort of opt-out. They just take whatever they want, and all’s well.

So that’s the playing field that we have today. Everything on the side of the companies, and the other side, the creators’ side, has empty promises and a hope and a dream. And so that’s where we find ourselves. Hopefully Nightshade can tip the balance just a teeny bit back.

BR: I’m seeing others trying to do something similar. There’s a startup called kin.art for example, that took a bit of a shot at you guys, suggesting you don’t go far enough and their approach is better. How would you characterize what they’re trying to do?

BZ: I can’t say that I fully understand what they’re doing, because it’s all private. All I’ve seen is a couple of superficial press releases. Until there’s some demonstration of real technical work, it’s very difficult to say anything. It sounds like they’re going to make it harder for companies to take the images and associate the right labels with them—[this is called] fuzzy labels.

The problem is that this is not what the companies care about—they don’t care about labels or alt tags. They just strip it all away. They use tools like Blip to automatically generate the prompts. Blip is a classifier that basically looks at images and tells you what’s inside—to the best of its ability. So if you want to give an image a fuzzy label that says, here’s a picture of a dog but it attaches the word cat to it instead. Okay, great. But the [image scraping] companies will just say, ‘Yeah, this website does this weird thing with the labels. We’ll just take all their images, strip out all the tags and run everything through Blip. Done.’

The other thing they mention is image segmentation, which is unfortunate because image segmentation is a classical term in computer vision that’s been around for decades. In computer vision, image segmentation just means taking an image and identifying the exact outline that allows you to extract a particular component of the image. I’m gonna guess that what Kin.art is doing is chopping an image into pieces and displaying an amalgam of those pieces on the screen. So visually that looks to you and me like a single image, but in terms of actual files it might be split into multiple pieces. Okay, so that might be an approach.

Except, again, the problem here is that this is a single web site. They’re not a moving target. So if I’m a company who wants to scrape their content, how hard would it be to say, ‘I know this is what they’re doing, and I’m going to run these images through a post processor that reconstructs the whole images from the partial images.’ Again, I’m inferring here. They could have interesting techniques, very complicated and very robust things that I don’t know about; this is entirely possible. I don’t want to put down anyone else’s approach or make statements about what they can and cannot do. I’m just saying these seem like very basic things. If it is that basic, then that’s not great. If they’re doing more complicated, smarter things, awesome.

BR: What struck me in their press release is the tone they took with your work, when you guys are the pioneers. So far all the coverage I’ve seen basically regurgitates the press release without digging deeper or asking tougher questions.

BZ: Which is really amusing—in their press release, they said Glaze and Nightshade work on the image after it’s been scraped. Boy, there’s a reason why we do that! I’ve been doing research in Internet security for more than 20 years. I’ve written scrapers and I’ve written scrape defenses. I know what that battlefield looks like. It’s a lost cause for a company, for a single website to say ‘We’re gonna protect your images from scrapers,’ when the scrapers are Microsoft and Google. Are you kidding me? Google has more IP addresses than small countries. They are running their own parallel version of the Internet that’s bigger than the physical Internet right now. And as a little website, you’re gonna say, ‘Yeah, we’re gonna stop all their bots.’ I mean... it’s good to be optimistic and idealistic. It’s just that I know exactly how these battles are fought and I can name 16 different strategies of how Google will easily get through any sort of anti-scraping technology.

So again, I wish them the best. I always find it counterproductive for people who are supposed to be on the same side to criticize each other. If their [approach] works, fantastic. It’s yet another thing to protect artists. I would still question their monetary business model. But having done what I’ve done for 20 plus years and knowing exactly how all these systems work, I wish them the best. I wish them good luck. It’s going to be a tough, tough fight.

BR: The other difference is that Glaze and Nightshade are free, whereas kin.art is charging a small fee for their platform.

BZ: You know, the one thing that is clear, I think… [sighs] I think this is not the time to be monetizing the sector. I think that a VC group that’s putting down $3 million to fund a company that says, ‘Let’s monetize off the sort of destruction that’s happening to the art community. Let’s monetize off the fears of these independent artists who are potentially losing their homes or not able to pay rent because they’ve lost their jobs,’ that’s just not a great business model. How do you convince people who are desperate for money to pony up money for commissions so that you can make returns on your VC investment? That part doesn’t quite work out for me and I think that’s just a misunderstanding of the art community and where people are right now.

BR: Your team and the kin.art team are all engineers, clearly. But there’s a marked difference in your approaches, and especially all of the advocacy you’ve done personally Ben, speaking on panels with Karla Ortiz and Jon Lam, and traveling and talking to artists’ organizations. Much respect.

BZ: It’s human empathy, right? Can you put yourself in someone else’s shoes—and if you can, you can actually be bothered to learn about how other people are living, and change your mind. But a lot of people don’t. It’s a lot easier not to.

And there’s so little profit margin. I see some of the numbers from the artists we talk to, and I’m like, God it’s brutal! You drive across the country to attend this convention, you sell this much stuff, you pay for the booth, the hotel, the food, and per hour, you end up maybe $10 an hour—IF they have a successful show. It’s just brutal.

So from day one, I said, we’re not monetizing this [Glaze and Nightshade]. It’s a stupid thing to do. These are people that need help—and you can’t get anybody to trust you if you’re trying to make a buck off them.

BR: I can attest to that personally, having organized—and funded—all of my daughter’s art shows and exhibits. Very tough to make a living this way unless you’re a big name. That’s why the unauthorized mass scraping of creative work has been so devastating for all these artists. So now that it’s been out in the world for a while, I mean Glaze, and now Nightshade, how is the broader world responding?

BZ: I’m active in the research community and I get to see a lot of papers under review. I will say there’s a lot more interest now in poisoning attacks. I’m seeing a lot of papers on poisoning and attacking large language models. I think lost somewhere in all this, are really good ideas and people who know much more than me in the space and who can probably do much more. So I’m hopeful that there are effective techniques out there that, once published, people will actually use and, as I said before, balance the playing field.

It’s funny, I was in Sweden in late January speaking to the Swedish Association of Artists, this big national organization, and one of my hosts had this very interesting take on things. He said, ‘I love this work because all systems need friction.’ In his view, Nightshade is adding friction to the system to try to balance things. That’s a very intellectual and healthy way to look at it. You can have a struggle and you can have a back and forth, but not unless the two sides both have something to work with. If one side has nothing, it’s really just a beatdown. So that’s what we’re trying to do, is attain that friction, that balance.

If you’d like to support the work Ben and his team are doing, get in touch. And be sure to stay subscribed to The Muse for more dives down the rabbit holes of the creative lives we humans lead.

Subscribe now

* If you know where these incantations come from, you are a true wizard. Or witch.

If you’re following the developments of generative AI (and its attendant AGI debates), by now you’ve probably seen the March 13 Wall Street Journal interview on YouTube (and the accompanying article) with Mira Murati, OpenAI’s CTO (and once upon two days CEO) about Sora, the text-to-video generation tool OpenAI has been teasing the world with.1

A few minutes into the video, Joanna asks Mira about the source of the video data used to train Sora:

Joanna Stern: What data was used to train Sora?

Mira Murati: We… used publicly available data and licensed data.

Joanna: So… videos on YouTube?

[Mira looks away, grimaces for three full seconds, then looks back at Joanna]

Mira: I’m actually not sure about that.

Joanna [sees she’ll have to pull teeth]: Okay. Videos from… Facebook, Instagram?

Mira: You know, if they were publicly available, umm… available… yeah, publicly available to use, there might be that data, but I’m not sure. I’m not confident about it.

Joanna: What about Shutterstock? I know you guys have a deal with them.

Mira: I’m just not gonna go into the details of the data that was… that was used. But it was publicly available or licensed data.

You’ve probably also seen, if not directly participated in, the questions and comments pouring forth from all corners of the Internet, to the tune of:

How does a CTO not know where the source data for their company’s products is coming from?

What’s Mira hiding?

Whoever media trained her did a pretty bad job.

OpenAI’s lawyers told her what to say. And what not to say.

Of course they used (stole) copyrighted videos. Otherwise she wouldn’t dodge the question like that.

OpenAI’s strategy is get to pole position no matter what, because by then the lawsuits will be too late.

More on all that here from the patiently prolific Gary Marcus.

In the meantime, let me tell you a story.

When I was in elementary school in what was then called Czechoslovakia, one particular third-grade morning still burns in my memory. The teacher was explaining to us the difference between capitalism and communism.

“Capitalism is bad because people don’t share things,” she said. “But communism is good, because people share things. It’s communal.”

I kept my pretty little blonde head still and my mouth shut. I knew better than to swallow the propaganda—yes, even at 8 years old. Many of us knew. We knew there was a world outside the Iron Curtain. But we also knew better than to speak out or argue with teachers. We kids had to sit in our seats with our hands behind our backs, nice and straight, and when the teacher entered the room, we’d all stand up and say, in well-mannered unison, “Good morning Comrade [teacher’s last name]!”

Hell we couldn’t even complain about dirt-encrusted potatoes in the grocery store. You never knew who within earshot was a clandestine informant for the Party. People get disappeared. Jailed. Tortured. And worse. You think cancellation is bad. Try risking arrest for trying to buy Kellogg’s Honey Smacks on the black market.

But this isn’t an essay about capitalism vs communism, or even about freedom (I have hella lot to say on that topic, coming from a country that had none, and also plenty to say about the two socioeconomic systems, having lived in both. Hint: neither one is all good, and neither one is all bad.).

This is an essay about trust.

Confido sed cognoscere

It is natural to trust. Trust brings safety, security, comfort, stability. In its absence, uncertainty, insecurity, stress, and anxiety come flooding in. Your sense of balance and stability start to liquefy.

Functionally speaking, it is easier to trust than to mistrust (or distrust).2 When you trust, you don’t have to research, fact check, or second guess. It’s face value currency. Mistrusting means you get invited to dinner but you’re the one who has to do the work: not only do you have to run out for groceries, find the recipe and cook, but you’re doing all the dishes afterwards while the person who invited you over sits on the couch with a margarita. Distrust means you know the cook is trying to poison you—and you still have to make your own dinner.

You might be shaking your head. I don’t just blindly trust random people. Or the mainstream media. Or politicians. What do you mean it’s easier??

Whether you arrive at the place of trust intuitively or as a result of research, or whether you get there instantly or slowly over time is irrelevant to this point. The act of trusting feels good, comforting, easy. The act of mis- or distrusting carries a lot of weight—and work—with it.

And yet, we need both trust and mistrust for our society to function. We need to be able to trust that every driver will stop at a red light, but still take that extra second or two after our light turns green. We should be able to trust that planes are still the safest mode of travel, but maybe follow the news about Boeing (and its entire corporate history) a little more closely—and not park in long-term parking beneath flight paths. We want to trust that the chocolate we buy isn’t linked to slave labor, even if we know it depends on what brand we’re buying. Happily, we can place 100% of our trust, completely blind, in the laws of the universe—time, gravity, and light move according to immutable standards. Until of course we get to quantum physics, but those are devils dressed up in details.

A certain level of distrust is healthy. You don’t automatically trust a phone number you don’t recognize. You don’t eat random berries in the forest. And you certainly don’t trust a stranger on the street asking you to drive them somewhere.

So why do we trust politicians who say what they know we want to hear, and do what they want once they’re in office?

Why do we trust CEOs? Why do we trust the media? The clergy? Our teachers and doctors and police officers? Why do we trust any authority figure?

Because we want a world we can count on. Strangers asking for a ride and strangely colored berries in a forest present potential danger; the people occupying important positions in society are supposed to provide for and protect us.

The unfortunate truth is that numerous public figures, particularly in business and government, provide for and protect themselves and/or their families and friends, too often at the public’s expense. As a result, we don’t trust them as much as we used to. A recent study by Gallup and the Knight Foundation found that half the respondents (close to 5,600 people) believe news organizations “deliberately mislead” the public. Fewer than two in ten Americans now trust the Federal government, according to the latest figures from the Pew Research Center (in 1958, when the study began, 75% of us trusted the government. We’ve come a long way, baby—a long way down). In a curious twist of irony, while business in general is seen as a more trustworthy institution than government (and that’s worldwide), individual executives (read: CEOs) are among the least trusted organizational leaders—this according to the 2023 Trust Barometer by Edelman. Want a double serving of irony? Take a peek behind the curtain of Edelman’s own client roster—sparkling greenwashed clean.

Speaking of irony. Being born in a country where political propaganda gets mixed into baby food either helps you build a strong and resolute psyche, or turns you into a cookie cutter pawn. You learn from a very young age certain institutions and figures are not to be trusted; you’re mistrustful by default. If you’re born in the land of the (supposedly) democratic and free, your default settings are set to ‘trust,’ and this gives you some catching up to do as far as trust in public figures and institutions is concerned—but it doesn’t take much these days.

It also doesn’t take much to surprise a friend or a loved one with a gift subscription—and support The Muse.

Give a gift subscription

Back to Mira and Joanna

In a recent conversation I had about the WSJ/OpenAI interview with a longtime friend of mine, a seasoned and successful engineer and entrepreneur, we explored possible reasons why Mira didn’t simply answer the question about the videos that were fed into Sora. It’s clearer than a sunlit day that there’s a deeper reason for Mira’s insistent refusal to answer Joanna’s questions.

Watch the full interview here (the data source questions start at 04:24):

OpenAI is in a tough spot, largely of its own making; Mira happens to be the person in the hot seat in this interview. She’s got a few choices:

A. Admit where the training material for Sora comes from. This would reveal OpenAI’s sources and potentially open the company up for legal action.3

B. Feign unfamiliarity with the source of the training data. This would drive public perception of her as an incompetent CTO—or a dishonest one.

C. Implicitly accept familiarity with the data sources but refuse to disclose them. This would avoid negative public perception of her personally, but cast OpenAI the company in a questionable light.

In the interview with Joanna, she exercises option B, effectively sacrificing the competence and integrity of her public persona for the sake of the company.

A Chief Technology Officer absolutely SHOULD know where the source data or material for their company’s products are coming from. In fact, Mira is a highly seasoned technologist and an experienced product lead—she could not hold the position of CTO in an AI company otherwise. Satya Nadella of Microsoft has sung her praises; Joanna Stern herself has previously outlined Mira’s experience and responsibilities in a November 18, 2023 article published by the Wall Street Journal. This is particularly telling [italics mine]:

“Murati’s teams are responsible for training OpenAI’s current and future large language models, making her an essential part of the decisions of which training data is—and isn’t—worth including and paying for. ... Murati has also spoken about the importance of improving data attribution to let people know where AI answers come from.”

Mira inadvertently contradicts herself in the video interview above. At first she tells Joanna she’s “not sure” where the data comes from, but finally takes a firmer stance and says “it was publicly available or licensed data.”

So, case pretty much closed. The CTO of OpenAI is directly involved in making the decisions regarding training data and its sources—as she should be.

Given all that OpenAI has done and been through, from infuriating creators whose content has been used for training LLMs to whipping up frothing global fear about AGI, there should be no surprise that it’s backed itself into a corner—and its leadership with it. If Mira had said, simply and plainly, “I’m not authorized to disclose that information,” without hedging or hesitating as she does in the interview, she would avoid accusations of incompetence or disingenuousness, but she would have shifted the implicit blame onto OpenAI—the underlying message would be not that she doesn’t know but that the company is barring her from disclosing the information publicly. We don’t know whether it was her choice at the spur of the moment, or whether she was directed to answer as she did, perhaps by OpenAI’s legal counsel. Either way, lately both the company and its executives have been shedding the public’s trust by the bushel.

It takes just minutes to wreck trust it took years to build up.

The primary problem here is that OpenAI, my friend the entrepreneur observed, has been deified. They’re the first company most of us think of when we think or talk about generative AI (largely due to this next sentence »). They’re the most immediately visible company in media coverage of generative AI; the media report on—and we eagerly gobble up—their every move, their every word and every tweet. The company is led by a virulently ambitious CEO who’s trying to have it all ways. As in politics, so in business: the more influence and money you amass, the lesser your ability to gauge your own fallibility and the deficiencies of your power.

OpenAI has the choice and the freedom to come clean, with all of us. No one is pinning Sam against the wall, threatening to do him in if he does right by all the people and organizations whose work has been used to train his models. The risk, which leadership is all too aware of, is the potential demise of the firm, through lawsuits and flight of capital. No doubt they would take down a few of their competitors as well, by association. It’s certainly a tough call, and it shouldn’t be surprising to anyone that OpenAI is doubling down. But there is also nothing stopping its leadership from starting over, and rebuilding the trust they should have focused on from the start—with creators, with the media, with the public. Perhaps this would start an unprecedented sea change in the tech sector, that their own power and fame are blinding them to.

No matter how much magical thinking Silicon Valley spins up, the ancient myths that underlie so many of the fall-from-grace stories we have seen over the years, from Theranos to FTX, are as steadfast as the laws of physics… we all know what happens to Icarus when he flies a little too close to the sun.

What few leaders seem to want to see, and what the myths are patiently waiting for us to realize, is that there is nothing inherently wrong with power, influence, or money. What matters is how you use it.

Do you know the password?

Good, you may enter. Do you know the password for the House?

I’m not sure… Is that a trick question?

Maybe.

There are still people in the world who have not heard about ChatGPT. Or Claude. Or Bard. Or AI in general. And quite frankly, my dear, they don’t give a damn. They’re trying to survive daily life, political oppression, wars, natural catastrophes, lack of access to water, food, health care, education, and often shelter.

Yet, if they are connected online, more than likely they—and the rest of us—are using some form of AI without realizing it. All it takes is a smart device and access to the internet. Most of us know the password for admittance, but we don’t know there is, in fact, no password for the House. And one day passwordless authentication might be an offer we won’t be able to refuse. AI is baked in—and the masks it wears are numerous and multivariate: social media, voice assistants, facial recognition, recommendation engines, smart thermostats. Anything to make modern life easier, smoother, less manual. Why walk across the room when you can simply exercise your vocal cords?

As a matter of principle, there is nothing wrong with making life easier. No reason to insist we continue to use manual typewriters instead of word processing programs (the frustration of having to retype an entire page because the typewriter ran out of correction tape is embedded in my middle school bones). No reason to scrub our fingers bloody when we have washing machines, walk or ride horses when we have cars, tend our own vegetable plots when we have grocery stores (or food delivery), spend three weeks on a ship when we can fly.

But the dinner party turns into a cult orgy when technology ceases to serve and support, and becomes its own end for profit and power.

The dark irony is that it’s the people with the least knowledge or awareness of AI who stand to be negatively impacted the most by its ubiquity: those struggling to survive in economies and political systems designed to stand on their backs forever; those not actively fighting generative AI companies in court for IP theft because they don’t realize their non creative jobs are on the line too; the people who committed the crime of being born into the wrong half of humanity, skin color, or socioeconomic status; and, most gutting, our children whose careers haven’t yet been conceived, never mind prepared for.

Those of us with the privilege and luxury to discuss whether our chatbots are conscious, if AGI will kill us or keep us, or which artist’s style we’d like to mimic today, are all too aware of AI systems, but our eyes are so wide open we see less than those whose voices AI is suppressing.

We’re having the wrong conversation.

Read more

Earlier this summer, of the well-regarded newsletter invited me to write a guest post. I had always found him to be thoughtful and supportive of the writer community here on Substack, and gladly accepted the invitation. We iterated on various angles to a few different topics, all of which were worthy and valid and intriguing. Ultimately, I chose the one I feel is fundamentally responsible for the tears and cracks in our social fabric: what I call the datafication of the human being.

The post is titled The Great Disconnect and it talks about the fragmentation, atomization, and loneliness that algorithms, be they of the social media or the generative kind, bring to bear on us living breathing human beings. I’ve reposted it here below for easy reference, or you can read it on AI Supremacy—and while you’re there, do check out other posts by Michael or his other guest writers. What I would like to do here is give a little more context for the reasons I wrote this piece and how it almost gutted me.

In 1988, Noam Chomsky and Edward Herman wrote a book titled Manufacturing Consent. This was shortly after the Digital Age ignited, but Chomsky’s and Herman’s work was prescient then and it is prescient now. If you haven’t heard of or read the book, please allow me to highly recommend it to you. It will give you a thought-provoking foundation for the kinds of conversations taking place today about the power and influence of social media and generative AI. The book’s Wikipedia page is illuminating—notably the part that talks about the apparently intentional demise of the publishing house that dared to publish the two men’s previous work, Counter-Revolutionary Violence: Bloodbaths in Fact & Propaganda. Gosh I cannot imagine what its parent company, Warner Communications Inc., was worried about.

Now that the masters of content have our consent, the next step is to atomize us, break us into little monetizable pieces that can be sliced and diced and tossed into shareholder salads. Our children are being born into a world that lives and dies by emojis, data points, and algorithms. We are being conditioned to train our own professional replacements; to filter our emotions, thoughts, and reactions into a set of neat, quantifiable, analyzable data sets, which then feed more content formulas back to us, and rinse and repeat; and in the process disassociate from our very minds, psyches, and bodies. We are being gutted into shells of our human selves. As so eloquently said in the comments on my guest post,

We are one of countless social species because this basic instinct allowed us to survive through various versions over millions of years. A social instinct has allowed us to find food, protect ourselves and procreate more effectively. We are wired for this at the DNA level - to communicate through affective resonance and pheromones, and to be physically touched. Lives in misalignment with basic genetic wiring become ill … and in this case empty/depressed, like Harlow’s monkeys.1

It took me three days to write The Great Disconnect, but the thinking and feeling that went into it took the better part of a decade. The moment I hit the first key, it was as if the fire department arrived with their pentagonal wrenches. It all came gushing out and tripped the breakers on my neurons. I had to stop multiple times—not because I didn’t know what to write (on the contrary, there was too much, and I had to cut it down to 25% of what I originally outlined), not because it’s violently complex (it is and it isn’t), but because the deeper I slid into the rabbit hole, the more clearly I began to see what was taking place at the molecular level. Out of respect for you, dear reader, I had to keep pulling myself out of the rabbit hole to get my bearings on the horizon, to breathe fresh air, and dive back in with a full tank of oxygen.

But this rabbit hole of technological isolation isn’t a simple straightforward tunnel that connects from the human world above to some deep dark tech cave down below. It’s a network of threads and pathways, of data points and algorithms, of pixels and patterns. It mimics our expressions, words, and actions; it mirrors our most extreme tendencies and ambitions; it amplifies our fears and anxieties. Like the vast tapestries of mycelia that weave our soils into the cradles of ecosystems, the algorithms that underpin the levers of modern society now run too deep and too wide for us to simply cut them at the roots.

Yet unlike the natural world, this algorithmic network has no life and no consciousness… no relationship to itself and no capacity to love, fear, or hate; it neither dreams nor wakes. It serves only those who know how to thread its circuits. In the end, it abdicates all responsibility because we are the ones who’ve built it, and we are the ones who are feeding it. It would behoove us to consider carefully where our willingness to feed this beast comes from, for we do so at our own ultimate risk.

A special announcement

Lastly, I’d like to shed some light on my MIA-ness since the last article I published a month ago, which was’s own thoughtful guest post here on The Muse. Aside from all of the usual and daily responsibilities a working mama needs to fulfill, without fail and often without recognition, the past few weeks have been spent preparing a brand new and absolutely delicious new Substack, the twin sister to The Muse. We go live with an initial soft launch here for the Substack community tomorrow, Friday, September 22. Watch this space!—and if you’re not subscribed, do, if only to meet The Muse’s twin sister 😍.

Subscribe now

And now, without further ado, here is The Great Disconnect. Settle in with a nice big cup of something piping hot and maybe a pastry… it’s a long read, and you’ll need the sugar to counteract this bitter melon.

“The Great Disconnect,” as originally published on

You know when the Surgeon General writes an OpEd about loneliness in The New York Times, things have taken on the color of ants, as they say in Latin America.

It’s not just the Surgeon General’s OpEd. Headlines all over the media ecosystem have been blaring about the mental health bomb of loneliness and isolation for several months now. It first blew up during COVID, when the world was shoved nose-first into a sea of N95 masks, everyone went online for everything, and cities turned into ghost towns. We disconnected in-person and reconnected online.

Much of that seems like a distant nightmare now. But not only has the COVID tide not fully receded, there’s a tsunami forming on the horizon that might make that quarantine feel like a kindergarten field trip. The robots are coming, the headlines now warn. AI is going to take not only your job, but your last few shreds of sanity and whatever connections and relationships you had with other humans, too.

​​During COVID, the skyrocketing ubiquity of social media algorithms in our lives turned into an overdose of epic proportions. Where social media dug a trench, generative AI is going nuclear. It has the potential to displace entire professional sectors; gut the creative class; re-thread our communal relationships with doctors, educators, and law enforcement; dissolve the nuances of romance; and turn our sense of what it means to function in human society inside out. 

But before we all run screaming for the hills, let’s unpack the thing that simmers at the core of AI’s power to disengage society: the Great Disconnect.  

Typing in prompts instead of working with your creative team to design a videogame, sexting with a bot instead of a hot human, or scrolling through digital galleries of beach sunsets instead of throwing your camp gear in the car seems like the perfect way to dilute life into a plate of cold broth. If you’re concerned ChatGPT and its bot friends might uproot your life and career at least a little, you wouldn’t be wrong. But you wouldn’t win debate class, either. The visible spectrum isn’t just black & white, there are more than a few ways to peel an apple (I feel you cat people!), and the enigma of superposition isn’t limited to quantum physics. 

What isolates and depresses you, might empower and delight someone less able or fortunate. What fragments and disconnects in one scenario, might unify and bond in another. Perhaps gen AI’s true legacy will be to make us all realize just how disconnected we already are, how much worse it can get, and spur us to turn the loneliness epidemic around. But that’s only if we’re willing to a.) see it; b.) accept our responsibility to address it; and c.) addressing it.

Great, but why should I care?

Fair question. I don’t expect this essay to resonate with everyone. But if you’re a startup founder, CEO, or manager in the process of integrating AI into your operations, you might want to know how to keep your people motivated, performing at the top of their game, and feel respected. If you’re a techie, wouldn’t you want to know how not to be rendered redundant, and why people are not doing happy hour anymore? And for the creators out there, as much outrage, frustration, and anxiety you might be feeling about generative AI, it is more strategic and useful to educate yourself about AI, how it works, where it fails, and how to ensure creators continue to thrive. Finally, if you’re any kind of human at all, unless you’re completely unplugged—in which case you’re probably not reading this post anyway—the worst thing you can do is dunk your head in the sands of denial. In the case of this formidable technology that’s poised to upend global society, ignorance is anything but bliss.

OK, but give me some context please

First off, let’s clarify the difference between loneliness and solitude. Solitude is the sense of contentment or even joy when being alone, present with one’s self. Those of us who write, paint, compose music, or engage in other activities that do not require communal participation, for example, often prefer solitude to do our work. I’ve never been able to join a writers’ group because I need to write alone. Loneliness, on the other hand, is a sense of isolation that persists whether or not there are other people around you. It’s that sense that something deep within is missing or out of tune, and perhaps even lack of a fulfilling inner life. 

Personal isolation and loneliness have been a focus of concern for the mental health sector for some time, certainly prior to the release of generative AI. There’s even a documentary film that bears the same title as this essay (I discovered it after deciding on the title). The Great Irony of the Great Disconnect is that the very technologies designed to connect more of us across the world, such as email, smartphones, and social media, did connect us but they also isolated and alienated us, each new technology driving us further and further apart. How can this be? Three critical shifts took place. 

First, the foundation of contact stopped being about communication, storytelling, and exchange, the way human societies have connected for millennia. It turned into likes, followers, heart icons, GIFs, and other digital compliments and feel-good indicators. Slowly, insidiously, our thoughts, opinions, and musings, and more crucially, our mistakes and unintended impulses began to be published for the entire world to see, and judge, and respond to, and, AND… reshaped as data. We’ll come back to this point, because it’s foundational to the discussion about generative AI. 

Second, the opinions of everyone reading our posts, listening to our songs and voices, and viewing our art and photos and videos, were given public life. Never before were creators—here I use the term “creators” in the most generic sense of the word—given such wide and deep access to the reactions and sentiments of the public, whether that public was their intended audience or not. 

Imagine what the Instagram feed of Cleopatra would have looked like! Assuming every inhabitant of Egypt had a connected device, the most she would have had is 2-4 million, a mere 0.95% of Beyoncé’s IG. Then again, Beyoncé doesn’t have her name carved in two-thousand-year-old sculptures.

Third, we began to connect at lightning speed with people we didn’t know in real life. Those new connections were no longer based on in-person, multi-year relationships, but digital representations of people’s lives, carefully—and not so carefully—curated. We went from getting to know the people we talk to over time and through experience, to making flash judgments on the basis of a single post or comment. As we’ll see, these three shifts have made a perfect storm of disconnection, alienation, and loneliness that generative AI will make quick work of if we’re not paying attention.

Datafication of the fully expressive human

The first shift in the way we communicate and connect, as discussed above, has to do with the mutation of the value and meaning of our expression. Digital communication tools, whether AI-driven or not, distill the words, images, and sound we express online into data. This data takes on various forms known as attention metrics, such as:

• Comments

• Likes/dislikes

• Icons for somewhat more nuanced reactions, such “insightful” or “funny”

• Shares & reposts

• Impressions

• Reach

• Open rates and click-through rates

• Bounce rates

• Length of time spent on a webpage

Of course, the full range of human emotions and expression does not fit into neat little categories, and that’s always a problem when you’re running analytics and trying to tie user activity to some kind of benchmark. It’s a particularly important challenge if you’re using those benchmarks to drive traffic to your online store, social platform, or search engine, or otherwise keep your investors well fed and happy. 

The biggest shift on the individual level is that we are no longer viewed as full human beings and bodies. We are “users.” “Consumers.” “Subscribers.” “Followers.” “Customers.” “Use cases.” Why not just say it out loud: we are Data Points. Stats. Numbers. In short, we have been datafied. Ironically, this datafication extends to everyone working in the tech sector as well, from investor to startup employee. They’re human too—and they post, comment, and browse just like the rest of the online world (albeit maybe a little less). This is the biggest step in the disconnect on an individual level: being human is an experience that is at once physically embodied, emotionally felt, and, for most people dare I say, spiritually traveled. When you consider yourself, or others, to serve the purpose of a data point, you’ve disconnected the humanity from the person.

Looking at it from the POV of the market, you need a way to quantify, analyze, and forecast the behavior, preferences, and opinions of all those annoyingly diverse and unique humans if you’re going to launch a (massively) profitable product or service. Well, you're in luck, because those annoyingly diverse and unique humans are also deeply tribal and constantly strive to fit in with their peers and communities, and so they’ve jumped right into the data pool, ​cold water be damned. We now routinely ask our readers to “please like, share, and comment below!” because we want data on them just as much as the algorithms do. The algorithms have trained us well.

Too well. The real reason we ask readers to like, share, and comment is not the data. It’s how the data makes us feel, and how easy the data makes it to get an instant result. It is a profoundly human thing to seek approval from our loved ones, friends, and now, increasingly, the rest of the world. It is an equally profound human thing to not want to have to do the work—the work of reading nuanced comments and opinions from fifty people, vs. a quick glance at that nice fat figure “42” next to the “like” icon atop your blog post. The number itself, of course, is relative—relative to how long you’ve been posting, relative to your overall email list size, relative to how many of your hard-earned dollars you’ve sunk into promoting that blog post. But it’s a concise, easy-to-grasp indicator of likability. What you miss, of course, is the real reason why any given person has “liked” your work. Some did it because they know and like you, and haven’t bothered to read the post. Others because they scanned the post and decided, in the spur of the moment, to like it. And some do it because they genuinely appreciated what you had to say. Now what about those readers who also read and appreciated your post, and maybe even told numerous friends about it, but chose not to ring that little bell? You’ll never know about them, just like Gabriel García Márquez never knew what the vast majority of his readers really thought of his work. 

And so the algorithms have effectively succeeded in transmuting our natural proclivity for peer and community approval into the data points they can work with. The invention of the “like” icon was a brilliant, if brutally world-altering, idea. Too many of us now create “content” with the intention of “engaging eyeballs.” (Whatever happened to the rest of our bodies? The disconnect thus becomes digital dismemberment, too.) 

How long—and deep—can we play the numbers game?

But why do we chase the highest numbers, instead of being content with whatever followers we’ve got? Because we can. Because the algorithms connect all of us in fractions of seconds. And because superlatives always impress: the fastest car in the Indy 500, the richest person in the world, the biggest predators on land and in the seas. The fastest growing app on the Internet (hello ChatGPT—no wait. Hello Threads, good-bye ChatGPT). So yes, we do numbers because assessing quantity is truckloads easier than trying to gauge quality and craft, and because we’re hard wired to seek and value resources, be they food, shelter, money, or yes, even those digital thumbs-up. In the latter case, digital status is a type of resource—it feeds not just our ego but expands the potential to earn revenue. (How some of us humans plan for that revenue is another matter.)

We can appreciate big numbers but we can’t process them properly beyond a certain size. Close your eyes right now and imagine a hundred-dollar bill. Now imagine a suitcase full of them. An entire warehouse of suitcases. How about a trillion hundred-dollar bills? Not so easy now, is it. Beyoncé can’t physically interact with every one of her 317 million Instagram followers either. I don’t pretend to sit in her head, but I imagine she doesn’t have the time to read all of their messages. 317 million does not represent the singer’s friends and peers; it’s a number that represents her success, her status on the world stage. Yet for all intents and purposes, Beyoncé is disconnected from her adoring tribe until and unless she pulls a few of them up on stage with her and puts her arm around them as they swoon. But what about the fans, from where they sit? How connected to their queen does any one of them really feel? Clearly, many do, as evidenced by their fierce loyalty and hordes of bee emojis. But is adoration true connection? And are the algorithms good for Beyoncé’s own mental health? Numerous sources would say not so much.

AI systems, on the other hand, can crunch unimaginable volumes of data, and they can do it 24x7. All they need is lots of processing power and a nice big data center with a steady supply of energy and water, the environment be damned. They are also completely unaffected by any of those pesky human emotional, psychological, or physiological consequences of excessive screen time or the nature of the content their own code serves up. 

When you, a human, spend even just an hour or two scrolling through the output of AI-driven social media feeds—TikTok videos, LinkedIn posts, Twitter (X) threads, how much mental or emotional energy do you have left for you, your work, and the people in your inner circles? How do you feel after subconsciously comparing yourself and your life—personal, professional, financial—to what the screen dangles in front of you?

And if you’re not impressed with the big numbers on Beyoncé’s Instagram, try the 500 million tweets posted daily on the site formerly known as Twitter, and the herculean task assigned to its recommendation algorithm to distill them down to “a handful of top Tweets… [on your] For You timeline.” It involves a neural network of roughly 48 million parameters “continuously trained on Tweet interactions.” Such work is not for the human of brains. Here’s a fascinating rundown of Twitter’s distillery—er, recommendation process if you’d like to slide down that rabbit hole.

Whether and how to connect, that is the question

Many argue that connecting online is still connecting—and in many respects that is true. I have personally made numerous connections online that have turned into friendships, including here on Substack. In some ways, the quirks of AI algorithms seem similar to the quirks of everyday life—a chance meeting in the hallway, on the train, or in line at a coffee shop seems just as accidental as stumbling across a thoughtful comment you decide to chime in on and end up with a new colleague or friend. And just like in real life, the experience and impact of those connections depend on the intentions and character of both parties. If you’re looking for trouble, you can find it online as well as in the real world. Granted, you can’t hide behind an avatar in the real world. At least not yet.

The dynamic of connecting with total strangers in an online world, itself disconnected from the physical environment, and interacting with them through text, images, and symbols, has inverted our natural, innate forms of communication. One could say we’ve evolved. A more accurate assessment, however, is that our own technology precedes us. We certainly know how to use it—and misuse it—but we remain biologically wired for physical contact and in-person interaction. You don’t need a stack of research studies to prove this (but here’s one, and another, if you like)—just notice how you feel in your body the next time you meet a good friend for dinner, vs. chatting with that same friend via text. 

The problem we have is not a simple question of quantity vs quality, or even AI vs human.  The problem we have is one of nuance, empathy, and context. Everything we do online is getting reduced to ones and zeros, to algorithms following instructions designed to maximize the indicators of popularity and profit. Influencers are voicing their frustration about their content being served less to their own fans and more to random people based on algorithmic assessments of what content is likely to spark the strongest reactions from which user accounts. When those reactions are less than pleasant, that can impact the influencer’s mental health. In fact, we’re seeing rising numbers of celebrities speaking out about their mental health struggles fomented by social media, and that is helping to bring the topic to the fore of public awareness.

Civil debate and thoughtful conversations do still exist online, but they tend to be drowned out in an increasingly virulent online environment that might get worse before it gets better. The feelings of loneliness, isolation, and the ever-present FOMO (Fear Of Missing Out) that social media has wrought upon us are being taken to a whole new level by AI-powered algorithms and the automated firehose of content and targeted recommendations they’ve turned on. It’s one thing to feel pangs of envy about the Instagram influencer with the crazy-perfect dream house, but it’s another to stare down the black hole of your entire professional career in customer service going up in flames because it is likely to be made redundant by a much more cost-efficient chatbot. And it is quite another to have law enforcement break down your door and arrest you while pregnant and getting your kids ready for school because their facial recognition system glitched.

Social media platforms do not interact with you; they are platforms, digital canvases upon which you, along with other humans, paint your words and emojis and little dancing pig memes. AI-powered chatbots are different—they talk back. They effectively step out of the platform and onto it, wielding their own talking brushes, as it were. Up until recently, any time we received a digital communication, we could safely assume a human initiated and produced that communication. Now that chatbots have blipped onto the digital scene, that safety has begun to slip away. We’re aware bots write conversational threads, and that’s primarily because of a.) their still-stilted communication style and b.) the context of their delivery: a ChatGPT window, a chatbot on a company website, a Siri on your phone. Would you be able to tell, assuming the chatbot’s natural language functionality is advanced enough, whether any given Discord, LinkedIn, or Facebook post or comment was written by a human or a bot? Maybe you can now; just give it a few months.

If certain interests have their way with generative AI, there will soon come a time when you won’t be able to tell whether any text or voice on any platform has been produced by a human or a bot, unless you have an established relationship with the human. 

The nuclear option

Since AI algorithms in general have been running beneath social media for a while, you might think this type of AI isn’t much different from other digital communication tools in terms of its impact on our isolation factor. But there is one simple but fundamental distinction, and that is the virulent, adaptive, and weaponizable nature of generative AI that other AI algorithms do not share. Specifically: 

• The ability to respond to your question or prompt in natural language (read: the ability to respond, period)

• The ability to produce new content in response to said question or prompt, in multiple modalities

• Training on billions of words, images, sounds, strings of code, and other content

• The ability to develop synthetic data

• The ability to analyze complex or voluminous data and find hidden patterns and trends

• The ability to automate and accelerate a broad range of tasks and processes

This kind of power and speed should not be taken lightly; in fact, it has been argued that generative AI systems like ChatGPT should not (yet) have been made widely available to the public. A little too late for that now.

Content, the meat and bones of the online ecosystem, used to require human energy and effort. Teams of copywriters, SEO analysts, marketers, social media consultants, and product managers would spend weeks building marketing and media campaigns, launching them, and gathering the results. Now we’re seeing teams of 4 accomplish the work of 3x that number with generative AI tools and workflows. This is an extraordinary improvement in efficiency of time, cost, and scale. 

So what do we do now

The AI cat’s fully out of the bag. ChatGPT & co aren’t the first AI interfaces to have impacted our lives, but this is the first time an AI system has captured the imagination and dread of the public in such profound and powerful ways. The Internet has been trawled for content gems without anyone’s permission, and literal tons of collateral damage float in the wake of the gen AI ship. Myriad questions and quandaries abound, from the professional to the very personal:

• Do you hire a designer or use Midjourney for your blog post images?

• Can you let that staff writer whose work you’ve always disliked, finally go and use Jasper instead? 

• Will you get sued if you generate a sound-alike song for your video?

• Can you trust code generated by AI? 

• How do we define (and enforce) copyright now?

• How can you tell an AI-generated dating profile from a real one?

• Should I trust the voicemails I receive are really from my friends and family?

• Is it ok if I want ChatGPT to be my BFF?

True to form, it’s the headlines that scream end of days that draw those clicks and shares, but the real issues are the ones with the quieter headlines. We need to worry about human misuse of AI systems much more than any AI-spawned apocalypse or conquest. Perhaps the most sobering reality that we will need to come to grips with is not what AI can do against us, but how humans can use AI to our mutual detriment, whether intentionally or unwittingly. Never has technology saddled humanity with such existential burdens. To borrow from the poet Robert Frost, the only way out is through the AI fire.

A few tips for the road: 

• Don’t outsource your creative or critical thinking agency to an AI system. Continue to rely on your experience, insight, and wisdom.

• Always put your people first. If you want to leverage AI in your company, identify the workflows where it can best support, rather than replace, your staff.

• Give them agency: involve your team in determining the best ways to integrate AI tools into your operations. 

• Don’t be seduced by the temptation of fast, cheap image generators. There is no replacement for skilled human artists.

• Likewise, don’t be seduced by what reads and sounds like a sentient entity. There is no mind, heart, or soul behind your chatbot screen.

• Never, ever forget there’s always a non-zero chance your AI dreams of electric horses.

Oh, and one more thing.

Thank you for reading. If you’d like to leave a comment, I’d be thrilled to respond.

~ Birgitte

This guest post is part of the virtual book tour for The Artist’s Mind: The Creative Lives and Mental Health of Famous Artists by my fellow Substack author , who writes about the complex relationship between art and mental health. Stops on the tour include an author interview on Sue Clancy’s website, a behind-the-scenes guest post on Great Books + Great Minds, and a book excerpt on Socratic Psychiatrist.

What a lovely way to get to know our fellow writers, explore the Substack ecosystem, and dive down a few new rabbit holes, wouldn’t you say??

Let’s turn it over to Kathryn:

As I sit down to write a post about how generative AI is impacting the well-being of artists, I find myself turning to a paragraph written right here on The Muse:

“We stand at a crossroads. It may, in fact, be a cliff. Some of us will sprout wings and chart new flight paths throughout the vast chasm that this fundamental revolution in technology has carved. Some of us will fall to our professional and artistic demise. Some of us will teeter on the edge, thinking it perhaps best to turn around rather than to tempt the Techno Fates—and gravity.”

Some artists aren’t at all concerned about AI and are even excited about it. Some artists are paralyzed with fear. Most of us are somewhere in the middle. If you’re teetering, you might find that anxiety pushes you over a negative edge. Let’s discuss what causes it, what it might look like, and what you can do about it.

Why generative AI is causing artists anxiety

There is a lot of fear, often warranted, about the potential impact of generative AI on artists and their careers. Just a few of those fears include:

• Losing artistic identity, authenticity, and originality, whether you choose to incorporate AI into your work or not. This can manifest as feelings of depersonalization or dissociation, or even devastation if you’ve suffered any kind of material or financial impact as a result of generative AI training on your art or undermining your ability to earn a living in your field.

• A sense of forced usage of and dependence on generative AI technologies. The message we keep hearing in the media is that artists need to “adapt or die.”

• Job displacement. As an artist, will you be able to make a living if AI can churn out digital images and other visuals much faster and cheaper than you can?

• Fear of your original work being stolen and utilized by others. This can range from a valid fear to debilitating paranoia.

Fear can create / exacerbate symptoms of anxiety, depression, and other mental health challenges. Artists who are mired in these or other fears about generative AI may find themselves in turmoil right now, because of the technology itself but also because of the media’s promotion of the idea that AI is taking over. 

Symptoms of AI anxiety

Anxiety manifests differently for everyone but some of the ways it may show up for artists specifically include:

• Constantly thinking about, reading about, and worrying about the impact of AI on their creativity and career

• Restlessness, trouble sleeping, and a desire to create but an inability to concentrate enough to do so

• Isolation and withdrawal from others due to anxiety and/or feelings of paranoia

• Rumination, fatigue, low self-esteem, hopelessness and other symptoms of depression

• Hypersensitivity to information, assuming the worst, hiding their work

Those are just a few examples. Trust yourself … if you suspect that worrying about AI is causing you anxiety, then there’s probably something to that.

Tips for artists to reduce AI anxiety

The number one thing that most artists can do to reduce this anxiety is to reduce the amount of time spent taking in information about it and use that time to create art. Focusing on creativity usually helps.

Here are some additional tips:

• Read up on all of the great arguments that exist for why human creativity is always going to exist and matter.

• Learn how to use AI tools. This will demystify them. You might find that you can even use them in your art (for brainstorming, for busting through creative blocks.) However, you don’t ever have to use them in your art to benefit from actually knowing what they can do.

• Practice holistic self-care. Nourish yourself with good food, good books, good conversations. 

• Practice mindfulness. There are so many great ways to use art and creativity to bring your attention to the present moment. In the present moment, everything is actually fine; anxiety is worry about the future.

• Spend time with creative people. Artist groups that inspire you will move your attention away from AI fears.

• If you feel like your AI anxiety is damaging your creativity or your mental health, speak to a therapist, mentor, doctor, or other relevant professional.

• Take back control over your art and your career:

  • Implement tools to protect your work physically, such as Glaze —a software app that protects digital images against AI scraping

  • Opt out your work from LLM databases

  • Get involved in initiatives, actions, and organizations working toward restoring artists' creative agency and rights. There's a massive fight going on right now between artists and the AI companies, on the government and the business stage, and the danger is that if artists lose, all of humanity will lose. 

Finally, recognize that technology has caused fear among artists across all of history. For every Andy Warhol that embraced screen printing, there was an artist who never touched it and did just fine and an artist who was immobilized by fear of what it was doing to the art world and the world as a whole. Heck, when printed books first emerged, many feared that they would end the creativity and connection created with oral storytelling. Things are moving faster now but looking at this bigger picture can help reduce our anxiety about this moment. Of course, it’s not the same; the scale and speed of the technology today is different and therefore the impact is different. The concerns are valid. But we can look to the artists of the past who chose to use - or not use - the technology that emerged and explore how their approaches might apply to us. 

Seeing this larger picture is one of the things that can help us get back to our center, reclaim our agency, and feel less fear about all that is happening in this space. Staying connected to the artist community and getting involved in efforts and organizations working to keep artists' work, livelihoods, and careers protected against the negative impacts of gen AI usage will also aid in these efforts. It can feel like big forces are taking over and their path forward is inevitable; but if anyone can change the course of history, the artists with creative minds are the ones to lead the way. They always have.

I believe that we are all artists with mental health experiences. In The Artist’s Mind, I explore how art and mental health intersect differently for each of the famous artists in the book. Some, like Edvard Munch, believed that their creativity and mental health symptoms emerged from the same place. Some, like Agnes Martin, believed that they weren’t related at all. Most were in the middle, and would have likely responded as uniquely to AI issues in art today as they did to the issues of their own time.

~ Kathryn Vercillo

If you’ve read this far, thank you for listening. The inner lives of artists and other creatives have always been impacted by social, political, and historical forces, but perhaps never before has the level of pain, distress, and uncertainty been this acute or this profound—nor has it been caused in such a singular degree by technology.

Perhaps we can encourage Kathryn to write her next book specifically on the mental health impacts on artists specific to generative AI. There’s certainly more than enough material.

In the meantime, check out Kathryn’s Substack:

Subscribe to Create Me Free

And of course, if you’re not yet subscribed to The Muse, that is something that should be immediately rectified.

Subscribe now

Welcome to Part III of the AI at Work series here on The Muse. Part I started off innocuously enough, walking us through a variety of work-relevant applications of generative AI and highlighted the four primary ways they’re helping professionals in the workforce.

Having secured our trust (and confirmed our naiveté), Part II dragged us through the mire of the rancid underbelly of gen AI. If you joined us on that daring adventure and survived, congratulations. You are now a wiser, humbler, and more critically thinking if not desperately more nauseated human. If you missed it and enjoy doom-reading or need to prove your humanly prowess to your significant other, please walk on through this door. Do bring your spear.

Here in Part III, we follow the White AI Dragon into a land of uncertainty and shifting realities. A land where all that you had known and relied upon as a creator, thinker, or maker of things, such as copyright law, intellectual property protections, the concept of authorship, and the foundational elements of accuracy and reliability, shift along fresh fault lines drawn by heartless algorithms. A land whose peace-loving, nature-aligned natives have not been prepared for the ruthless colonizing onslaught now being visited upon them by technology so new they shake in terror and awe and consider it their own myths and prophecies incarnated.

This series lives behind the paywall, so I can feed the insatiable data dragons. But if financial constraints make it challenging for you to upgrade, drop me a line and we’ll work it out—especially if your profession is being impacted by generative AI.

Climate change is not enough. Threats to democracy are not enough. A global pandemic is not enough. Another war in Europe (and all the others around the world) is (are) not enough. A mental health crisis among our youth is not enough. Because we apparently have way too much free time on our hands, and our lives are far too comfortable and convenient for our own good, we also need another existential crisis: the (another?) AI revolution.

I’ve also had more than enough of the “unless you’ve had your head stuck in the sand the past few years, [fill in latest amazing insight no one else is writing about]” cliché intros to articles. But the AI dragons do burrow through the sand, so ironically, the best way to get to know them and how they work is precisely to stick your head in the dunes. Ready?

Read more

In Part I of my quickly ballooning1 “AI at Work” series, we walked through some of the ways people use generative AI at work, and highlighted the four primary ways professionals are benefiting from integrating gen AI into their work.

Here in Part II, we get down on our knees and crawl through the less savory aspects of gen AI. Bias. Creator compensation (or rather, lack thereof). The environmental footprint. Identity appropriation. Data labeling sweatshops. And more foul-smelling roses. As those familiar with dragons will tell you, there are benevolent dragons and there are malevolent dragons. And as with any world-changing technology released before its ripening time, things will break. Bones will shatter. Heads will spin. But there’s no other choice but to walk through the fire.

Alright, let’s roll up those sleeves and get to work shall we. There’s a lot of ground to cover. Er, slime-slicked AI dragon burrows to crawl through.

This series lives behind the paywall, so I can feed the insatiable data dragons. But if financial constraints make it challenging for you to upgrade, drop me a line and we’ll work it out—especially if your profession is being impacted by generative AI.

Read more

Generative AI—the algorithms that produce text, code, music lyrics, photo-like images, and digital art in response to user prompts—is disrupting numerous sectors throughout academia and business. Its rockstar, ChatGPT, gained over 100 million users in the space of the first two months following its launch. In the AI at Work series, I look at the ways professionals in a variety of fields are using (and not using) generative AI, and the associated risks and benefits of this technology for their profession. Before we dive into the first rabbit dragon hole, let’s take a first-principles look at some universal truths about the way generative AI is impacting, or has the potential to impact, the world of work. “Bubbles in the Sand” parts I – III lay down the foundation for this exploration.

These first three posts were free through August; they have now gone back under the sands, behind the paywall. Why? Because I need to feed the data dragons, and those beasts are perpetually ravenous (I had managed to hide these first three posts from them, but they’ve sniffed out my tracks). If financial constraints make it challenging for you to upgrade, drop me a line and we’ll work it out—especially if your profession is being impacted by generative AI.

From education and publishing to automotive and manufacturing, the worlds of work and industry are facing potentially unprecedented disruption by AI systems, specifically generative AI.1 Uncertainty and fear swirl around the impact generative AI is already making in some fields, parting the waters of the working world into two sides: those embracing AI with open arms, whether for good or ill, and those fervently opposed to its all-too-active misuse, calling for strong regulation and protections. And yet, the parted sea analogy is not quite accurate, for there is a third cohort: those who see both the real-world impacts of gen AI, be they constructive or destructive, and the very real need to rethink, redesign, and reshape the way we work—and not just because of an algorithm.

When you see bubbles popping up in the sand as the waves recede, you might think those are pockets of air trapped below by the churning current. And you’d be right… at least about the tiny bubbles peppering the length of the shore. Some of you, though, might have a better idea—it’s more likely the Pacific mole crab digging its way below the sand surface in the swash zone. In the wake of the economic tides crashing on the shores of our livelihoods and careers, though, the bubbles are neither tiny nor caused by little creatures. There be AI dragons under them thar sands.

Ready to dig a few dragon burrows with me?

Read more

In Part I of Hallucination Nation, we explored the meaning and relevance of three innocuous little words now causing so much havoc in the world of AI. We also vicariously watched me having a few all-natural, literal hallucinations.

Let us now continue with the discussion. Here in Part II, we turn to these mysterious AI hallucinations (read: errors or confabulations) and take them for a spin in the real world. If you haven’t read Part I yet, I strongly encourage you to do so—you must meet Mr. Error! He’s a sexy one.

By the way, if you find the literary reference in the title of this post too obvious for words, email it to me and you’ve got yourself a free 3-month paid subscription to The Muse. While inspiration lasts. Not subscribed yet? Whatever are you waiting for! Here below!

Subscribe now

A warming AI world

Somehow, in the blink of the Internet’s eye, we have gone from expecting a high level of journalistic integrity of our news media to slurping daily doses of propaganda and misinformation, and now to giggling over cute algorithmic gaps in data integrity.

Oh look, Charlie! How eminently adorable. Our little AI’s are hallucinating! Let’s sing them a lullaby... for when they awaken, surely they will reformulate their dreams into more accurate, fact-based, real responses. Or maybe we’ll just grow to love them with all of their curious and indelible flaws.

When a chat bot delivers inaccurate information or references to things that do not exist (scientific papers, news articles, the names of authors, etc.), what should our response be?

a.) Close the browser window and refuse to use the bot again until it can be relied upon.

b.) Accept it as TRUTH because we either don’t know any better or it does not occur to us to double-check the information

c.) Laugh it off as a “hallucination” and post screenshots of the exchange on social media

d.) Write an article about it, screenshots included, calling for the public to remember to fact-check the replies they receive (and for the AI developers to maybe retract their bots)

I’ve seen instances of each one. Personally, after numerous sessions with both ChatGPT and Google’s Bard, I closed the windows. There’s no doubt these technologies have tremendous potential, but you have to be clear on what you’re using them for and how. As far as search and research go, they’re not there yet—and it’s not my job to teach them to get there. I have yet to hear a convincing argument for becoming a glorified—and unpaid—bot babysitter or tutor. If I cannot place my trust in an algorithm when it comes to research or obtaining any sort of accurate, reliable information, it makes precisely 0.00 sense to continue using it. I’m not suggesting that everything online should be absolutely accurate or truthful—the dear lord knows how much chaff has been strewn around the Internet. But if you create a thing (product, service, tool, etc) intended to be used by humans for specific purposes, and you release that thing into the world, and you encourage people to use the thing, and the thing then fails to work well, is that not a valid reason for people to return the thing and get their money back? Let’s give this some real-world perspective.

Would you continue to buy lattes from Starbucks if they randomly started serving brown dish water instead of real coffee? Some days you’d get the coffee, but other days you’d get dish water. You would have to check each and every cup to make sure the “coffee” is real. Appetizing thought isn’t it.

Would you continue to drive a car whose front left wheel inexplicably fell off a few times each week? And sometimes it would turn into a set of stairs, which of course would be just weird. You would have to carry a tool kit with you, or at the very least a really good emergency roadside service because you’d never know when you’d need it. (And maybe a wizard’s handbook to turn the stairs back into a tire.)

Would you continue to hike up a mountain that randomly dropped out from under your feet, or suddenly raised itself up a couple hundred feet, knocking you off the trail? Too bad if you’re close to the edge…!

Would you continue to work with a colleague, date someone, or hire a babysitter—or engage in any sort of activity with another human who was randomly unreliable, spewed falsehoods and inaccuracies at random intervals, but who was almost always very polite about it?

Oh hang on, you say. I get the Starbucks analogy but you can’t compare a mountain to software. Mountains are made by Nature, and software is made by humans. Mountains don’t just disappear. They’re always there. They’re, you know, reliable.

Except when a cataclysmic earthquake cracks one of them open, or it turns out that mountain is a volcano that’s just woken up after a few million years of slumber, or there was no mountain there because you were having a lucid dream.

We can debate what is real, and what is really reliable, for a good solid week or so. Point is, a functioning society requires a certain foundation of fact, reliability, and accuracy. This is not a rant against AI. It is a call for AI developers to slow down and not break things. And not take credit for others’ work.

It’s this slow yet lightning-fast progression through evolving stages of misplaced trust, false assurances, corporate gaslighting, FOMO, a dash of pure unadulterated tech bro ego, and evaporating critical thought on a mass scale that has brought us to this point. We are the proverbial frogs in the pot, the fish in the ocean who don’t realize how wet—or hot—the water is (yes even the oceans are warming now). I doubt we would have accepted these so-called hallucinations quite so easily if we had not been taken for a walk (read: ride) through the Purgatory of Fake News these past several years.

Hallucinated Histories

Anything worth debating always needs its proper context and history. The history of human hallucination is one hell of a spiderweb—and I’m not referring to psychedelics. I’m referring to history, the way it has been written, spoken, recorded, and remade. Even today we’re squabbling over who gets to tell whose story—enough there for an entire book, but to give just one example: it seems that the history of the trade of human beings in the US is uncomfortable enough for some that they endeavor to deny it and ban books about it. Those are true stories that some try to suppress. A different matter is false stories that some strive to promote.

We humans have an extraordinary faculty for imagination, and imagining things. Whether those imagined things are right or wrong, good or bad, true or false, is beside the point. The point is, we are capable of imagining them, and expressing them. Our imagination is our gift. What we do with that imagination is our responsibility.

Is not fiction imagined? Is not fiction, well, fiction? We do not consider fiction bad, immoral, or wrong; we instinctively understand the purpose of fiction is to allow us to dream, to live other lives, to imagine places and people we’ve never seen or met. Fiction does not pretend to be real (usually). Non fiction, be it an essay like this, books, news, or documentaries, also carries a purpose, and that is to communicate facts, realities, observations of our world. It is when non fiction uses the sex appeal of fiction, puts on the makeup of artifice, and dresses up in an imagined story to fulfill an ulterior motive or agenda, that the trust between writer and reader, between creator and consumer, starts to break down. It can be innocent, but it can also be malevolent, and all manner of colors and textures in between.

All world religions have stories they share with their followers and believers. It is not for me to place value judgments on those stories, but let’s just say some are more fanciful than others. Governments, political groups, companies, and numerous other constructed groups all have their stories as well. Some of these stories have started wars (the Great Crusades), some have caused horrific mass murder (the Holocaust), and some have only recently been de-veiled (Christopher Columbus’ “discovery” of the Americas).

Then there is what we used to call yellow journalism. This article runs down a brief history of fake news—quite a read! How ironic, too, that Joseph Pulitzer, the famed newspaper publisher who lent his name to one of the most esteemed honors in journalism, engaged in rumor mongering as part of his ongoing rivalry with William Hearst. In other news, did you know that the original founders of YouTube apparently had no issue with videos that had been uploaded without the permission of their copyright owners? That inflated their stats of course, Google bought the company, and the rest is history—more crucially, the rest of their competitors, who played by the rules, faded away from that history. The alternate history that could have been, had they all played on the same ethical playing field. More recently, another promising and well-intentioned startup felt the sharp teeth of a predatory bigger fish.

Examples like these are like raindrops in a storm—hundreds of thousands if not millions. Fake news, false stories, tales taller than the Eiffel Tower abound in a dizzying array of form and channel, while honesty and fact are choked by a Sargasso Sea of deception. Guess what “content” made it into the training data sets. Is it any wonder that the hiccups and errors we now see the chat bots spit out, are equally if not more inaccurate, biased, and non sensical? GIGO,1 as they say.

What is most dangerous isn’t the stuff generative AI comes up with. It’s the fact we can no longer rely on our Fourth Estate, and perhaps we never fully could. Perhaps it’s to be expected that we would accept AI-generated confabulations in our lives after the constant diet of fake news and so-called fake news that were actually real, confusing the issue even further. If someone can manage to dissolve your mind, how do you put it back together?

Flowers for AI

Like the rodent character in the famous short story that inspired the headline for this essay, AI’s purported intelligence, crafted mechanically, algorithmically, artificially, is destined to decline as we feed it with ongoing cycles of our own laziness, willingness to surrender creative agency, and lack of critical thought. Our own desires for templatized entertainment, poems about electric vehicles in the style of Shakespeare, and plasticized movie scripts with synthetic AI voices. Our own willingness to let an algorithm plan our dinner menus, parties, weddings. Our own gullible fear of being left behind in the rush toward some prefabricated utopia, not realizing it’s the corporate power players who have bamboozled us into rushing headfirst for that cliff, giving freely and without limitations of our time, our energy, and our mental faculties to take their chat bots and AI interfaces to the next level, trusting their promises to give us more strategic positions and higher salaries. But at some convenient point they will pivot and performatively wring their hands and issue PR-peppered apologies saying oh actually we no longer need even the higher-skilled white collar workers, we’re good thanks!

It doesn’t have to be this way. We can choose to harness the powerful algorithms of AI to solve our most pressing problems. Let AI fold proteins as well as laundry. Let AI model climate predictions and good behavior in online comments. Let AI outwit the fraudsters, the spammers, the plagiarists, the criminal masterminds—and put them out of business forever (or at least make them buy their own merch).

But leave the artistry, the imagination, the inspiration, the creativity to us. With or without digital tools. AI doesn’t care what it does; it’s a piece of software. But we do. We care whether we will be able to earn a living doing what we love, and we care whether we’re able to choose how we create.

Given the global outcry by the creative community against the excesses of AI, it certainly won’t be an easy battle. It might be that the data-hungry, profit-motivated, power & status-obsessed players have finally met their match: the now-truly-starving artist. Just as that old saying about the cornered mouse proves to be stunningly false, so the starving creator can grow some gnarly teeth pretty quickly.

Author and activist Naomi Klein has written a well-thought-out opinion piece about AI hallucinations in The Guardian. Ms. Klein’s headline reads “AI machines aren’t ‘hallucinating’. But their makers are.” She outlines four “hallucinations” that the AI power players seem to be having, and cautions us against falling for them. But they’re not hallucinating Naomi. They’re all too clear-eyed.

One last thing. The Flowers for AI aren’t real. They only bloom backward into noise.

If you’ve gotten this far, you are either pumped up on a double espresso or you accidentally hit the “page end” key. Either way, I hope this has been a worthwhile ride! Part III of Hallucination Nation is in the works (can’t give a precise date because I haven’t integrated my writing calendar with ChatGPT, and have no intention to). In the meantime though, share your thoughts below and share The Muse with friends!

Subscribe now

Share The Muse

Lector cave… it is not recommended to read this post just prior to sleep. The author cannot be held responsible for the nature or quality of the resulting dreams.

“I see a cave with purple and bronze textures on the walls.”

“Wow really. Ok how about now?”

“Mmm now it’s green, like moss or something.”

I was lying on the chiropractic table, face up. She was pressing one thumb into the side of my ribs, and the other thumb somewhere along my leg. Each time a different set of pressure points—my head, the other leg, an arm, the hip. Each time, it triggered an image—as if she were playing me like a piano of color and texture. She was a graduate student at an acupressure institute in Los Angeles; I was a young screenwriter working production gigs on the side while I chased my big break. I had volunteered to be her guinea pig—I’d do anything for a free massage, and acupressure was the next best thing.

Except neither of us expected there’d be a light show. She’d never seen anything—anyone like me, apparently. She forgot all about what she was supposed to be doing and started trying out different pairs of pressure points to see what kind of visions each pair would trigger. We had a lot of fun that afternoon.

This wasn’t my first hallucinogenic trip.

Disclaimer: I have never, in my entire life, taken any mind-altering drugs, natural or synthetic, for any reason or purpose whatsoever. I have never even smoked a cigarette, much less weed or anything smokable. I’ve had a BBQ grill blow a gas flare in my face and my eyelashes singed by a candle, but that’s pretty much it. I still don’t know where the hallucinations came from.

A few years prior, I had moved to France for my junior year abroad. It was there that I started hallucinating these… images, these visions of color and texture. The most striking one was a large tunnel of fabric… a gigantic piece of white fabric with large red polka dots, swirling and twisting itself into a tube. It was even properly shaded, with light at the end of the tunnel (I guess my subconscious likes puns). The other visions were various textures and colors that would appear randomly — not every day, but enough throughout my year abroad that I certainly took notice. They weren’t dreams; they popped up randomly when I was fully awake and going about my day. My French friends thought it was all super-chouette, as they loved to say.

It did occur to me to maybe go get a medical opinion, but I didn’t have the money to get an MRI done or see an expert (too bad I didn’t know back then). The visions were, well, visual, not auditory (as in, those voices in your head), didn’t interfere in my daily life, and I certainly didn’t hurt myself—or anyone else—because of them. So, I left it alone. I figured it was my subconscious mind welcoming me back to Europe, the continent I had left as a child when my parents emigrated to the U.S. Either that or a really overactive subliminal imagination.

The experience with the acupressure student in Los Angeles was the last time I had these kinds of visions, and pretty much the last time I thought in any substantive way about the action of hallucinating.

Have you ever hallucinated anything? Please share in the comments!

The prodigal dream

How ironic it is, nearly three decades later, that hallucination comes knocking on my door again. Not in my head, this time, but from the outside. Peeking into the windows, slipping letters under the door, trying to convince me it’s different now. It’s all grown up and it’s got its own place. It has a serious role to play in society, a real job.

Dear reader, meet the AI hallucination.

Most of us have by now come across this term in the media and various Substacks. We have been made to understand that a hallucination in the context of chat bots like Google’s Bard or OpenAI’s ChatGPT is a response to a query or prompt that contains falsehoods, inaccuracies, or content that was created without any basis in factual information. There’s even a Wikipedia entry for it—it defines an AI hallucination as “a confident response by an AI that cannot be grounded in any of its training data.”

The very first time I saw the word hallucinate in an article about ChatGPT, I closed my eyes, cringed, and screamed Nooooo! silently inside. I knew it was too late to stop the anthropomorphism tsunami that had already started. And I could see the onslaught of inaccuracies and false information that would be flash-excused, en masse, by the public—faster than DALL•E could generate an image of melting clocks in the style of Dalí.

WARNING: dictionary definitions straight ahead 😱

Let’s revisit the basic definition of hallucination:

Hallucination

/hə-ˌlü-sə-ˈnā-shən /

noun

A sensory experience of something that does not exist outside the mind, caused by various physical and mental disorders,1 or by reaction to certain toxic substances,2 and usually manifested as visual or auditory images.

The sensation caused by a hallucinatory condition or the object or scene visualized.

A false notion, belief, or impression; illusion; delusion.

As you can see, the two definitions of the term hallucination are rather different. The dictionary definition is a “sensory experience,” a “sensation,” a perception of something that doesn’t exist. This requires some level of consciousness, awareness, a living organism with the capacity to feel and experience said hallucinations. The Wikipedia definition claims that an AI hallucination is merely a “response,” or output that is not grounded in facts or accuracy. Ok, so technically not a hallucination.

We can debate the evolution of language all you like, but let’s be very clear-headed about the selection of this particular word to describe the hiccups our AI friends are having. It is yet another instance of anthropomorphism, a tendency turned trend (and yes, I did just anthropomorphize AI in the previous sentence, intentionally). If you step back a bit from the overhyped media landscape, you’ll see article upon TV program upon YouTube video upon podcast upon more articles/TV shows/videos/podcasts about “interviews” and “conversations” with AI, as if it were some sort of sentient celebrity being. How is it we never talk about Adobe’s Photoshop or even Grammarly3 this way? Aren’t those also digital tools for creators?

Ya, I know, it’s because Photoshop and Grammarly don’t talk to us and they’re not buying us drinks. Yet.

A more accurate term for what is taking place behind the chat bot’s front end would be confabulation. AI researcher and author used the term recently in one of his on-air interviews4 (kisses for that Gary! 💋 ). Let’s take a look:

Confabulation

/ kən-ˌfa-byə-ˈlā-shən /

noun

The replacement of a gap in a person's memory by a falsification that they believe to be true.

According to the National Institutes of Health’s Library of Medicine, a confabulation is a “neuropsychiatric disorder wherein a patient generates a false memory without the intention of deceit.”5

Note the critical difference here, which is the conviction or assumption that the thing in question—whether image or information—is, or is not, real. In the case of a hallucination, the mind knows, on some level, that what it’s seeing isn’t really there. In the case of a confabulation, there is a gap in memory that the mind fills in with an idea or a memory it believes to be true.

If an AI system were to truly hallucinate, it would create an image or sound completely of its own volition, rather than a response to an external prompt. Anthropomorphic fantasies aside, when an LLM fills in a gap in the strings of predictive text that it spins off with a word or phrase that renders the output inaccurate or false, that’s not a hallucination. The LLM is not imagining something out of the blue. It’s responding to a question or prompt typed in by a human, and sometimes it hiccups and puts in words that don’t make sense or aren’t true or refer to things that don’t exist. So that would be a confabulation, then.

Ok, you say, so it sounds like we need to send out the big important READ THIS NOW OR LOSE HALF YOUR AUDIENCE press release to all the media outlets and have them execute a Find and Replace query. Replace “hallucination” with “confabulation.” Easy fix!

Before we turn on the presses though…

We’re anthropomorphizing again. The above definition of confabulation couches the concept in the context of the human mind, not code.6 Not software or hardware. In its most fundamentally basic, simple, foundational form, when an AI chat bot returns a false or inaccurate text string, it makes a mistake. An error.

This is a radical, daring thought, I realize. Perhaps best to sit down and take a few deep breaths here. Maybe light that special scented candle too (but don’t get too close your eyelashes might get singed). Ok ready? I have no choice but to say this part out loud: why not just call a spade a spade, and use the word… error.

Dear reader, meet Mr. Error. He has been buried underground since November 2022, his soul expunged from the collective consciousness of the modern Western world and its social media overlords. The chain he bears has but one key to release him, and it is you who hold it. Let us now resurrect thee, Sir Error, and dub thee Knight of Appropriate Phrasing! May you go forth and pry open the caked-over neurons of all those who hath sipped excesses of AI Kool-Aid! Rise, in error but good faith!

No, really, I have not smoked anything or eaten anything weird for dinner. It’s just 12:30 at night and this image of a mud-caked man in chains just spoke to me. Also, asked me what I’m working on and I said I couldn’t stop writing. It was then that I made the executive decision to split this essay into fraternal twins; Part II comes out next week. Back to Mr. Error, now that he’s been fully anthropo– god I’m sick to death of typing that impossible word— fully formed in the flesh.

Error

/ ˈe-rər /

noun

Something produced by mistake (like maybe the image above. I really hope he got paid, or at least a nice dinner, for being slathered in mud just so writers like me can paste the photo into their posts about hallucinations).

A deficiency or imperfection in structure or function.

An act involving an unintentional deviation from truth or accuracy.

Surely, calling a software error simply “error” is a profoundly unconventional idea that might destabilize entire paper skyscrapers that the media noosphere has built and glued together with the drool of a shocked and awed public. I shudder at the thought of Mr. Error walking the streets of our cities asking law-abiding citizens for the key to his chains. But there is hope—we can take comfort in the wisdom and experience of other industries and sectors that have been infected with the erratum virus—and lived to tell. Newspapers routinely publish errata & corrigenda. Economists, investment firms, and Wall Street sharks have been known to fudge a number or two (or most of them). Scientific research isn’t immune either. Hell, even Nobel laureates aren’t perfect.

All this to say…

Cogito ergo erro, n’est-ce pas? You can ask Bard or ChatGPT about that, but they might hiccup and throw an error if insufficient humility has been coded into their algorithms.

Ah, but error is not quite as romantic as hallucination, is it. Hallucination carries with it the connotation of an ethereal, dreamy state… imparting a halo around the imagined vision, as if it were lucid and light, lucid enough for the mind to believe. The very sound of the word lilts, breathes, sighs, as we swoon with it, losing ourselves in the deep ocean of a new, hypnotic intelligence that talks to us in natural language. Error grates against the tongue, the three r’s rubbing crushed gravel against our smoother sensibilities, and the short length of the word compacting any aesthetic-auditory appeal it might have had into the equivalent of a crushed empty can of soda, ready for the recycling bin. And error, embodied in the earthly (er, muddy) form of Sir Error, takes no prisoners. He still wants that key, and he knows you have it.

Part II of Hallucination Nation drops next week, where we give AI hallucinations a bit of a hard time. Because, the real world. We’ll see if I can dream up a Part III (am taking suggestions!).

Enjoyed this post? Hit the ❤️ button above or below (apparently it helps more people discover Substacks like this one and that’s a great thing). And sharing is hallucinating together with your friends. Or something.

Share

Last week a Note from popped up in my feed. She said:

AI optimists say that AI will never replace knowledge workers' jobs. I beg to differ. I am already hearing from fellow professional writers that they are being told to use ChatGPT to speed up their work or "make it more efficient" - whatever that means.

The problem is, if there's a way, people will try to use AI to reduce human capital.

I’ve been hearing this too. I replied:

This is precisely the issue. It’s not AI. It’s us humans. We’re like an auto-immune disease: we attack each other. Humans telling other humans to use AI to “speed up.” Do more. Be more efficient. Never mind your intelligence, your creativity, your insight, wisdom, integrity, brilliance. Just get us the data, the profits, the stats. And STAT!

To which Elizabeth responded “Yes. They are exasperated that humans need time to think, need rest...” Another writer, JD McBride of , reminded us that we humans also need treats. Which, if you think about it, is a critical point. Anything that talks but doesn’t need treats is highly suspect (hello chat bots). Our talking parrots (not the stochastic ones) would wholeheartedly agree, methinks.

By the way, this is how JD and I discovered we are chocolate soulmates. For those of you who love chocolate as much as I do, stay tuned. Boy do I have a treat for you later this year!

Our problem isn’t technology. It’s not robots. It’s not AI. Our problem is us. There is no malignant evil lurking in your CPU, watching you while you sleep or… do other enjoyable things at night. No software has suddenly metamorphosed its code into dark incantations of destruction against its human authors. The spam that’s turning your Inbox into the Spamgasso Sea? Humans are sending those.1

Who’s using ChatGPT to write their essays or homework and pass them off as their own? Who’s packaging up boxes of prompts and selling them for profit? Who’s inputting their company’s confidential information into the bot thinking they’ll have less work to do? Who’s using it to write pickup lines, personal emails, and ripping off dead playwrights?

Humans.

Who failed to ask permission of, and offer compensation to, the writers, artists, musicians, developers, and other creators of online content, before they firehose-fed the LLMs (Large Language Models) with billions of pages of content from the Web?

Humans.

Who released the LLMs before they were fully baked and set up with the proper guardrails?

Humans.

Who is now wringing their hands, leaving prestigious corporate roles, wishing they had seen this coming earlier? Who is—reluctantly perhaps—speeding up their own development and launch processes so they can compete with their arch nemesis and not lose market share?

Yep, you guessed it. Darn humans. (Remember corporations are people too.)

But let’s be fair. It’s not just the poor bots that are getting scapegoated as the villains of our latest obsessions. Think back to… oh late 2019. An innocent little virus called SARS-CoV-2 was carried by its human hosts on planes, trains, and automobiles to many other humans everywhere. Soon it was partying inside the bodies of people in just about every country on Earth.

Do you blame the virus or do you blame the humans who passed it on to other humans? Neither, or both? Maybe you blame the system that allowed the virus to get to that Patient Zero in the first place.

It’s not a coincidence we use the word virus in both biological and technological contexts.

WARNING: dictionary definition ahead 😱

Auto-immune disease

A condition in which the body's immune system mistakes its own healthy tissues as foreign and attacks them.2

I would define certain types of automation (e.g. Large Language Models) as a condition in which human society mistakes its own extraordinary talents and abilities as disposable and neutralizes them. Just like the macrophages, neutrophils, and lymphocytes of the immune system that misinterpret healthy cells as dangerous invaders, some authorities in human society (e.g. managers, VPs, CMOs, CEOs) are misinterpreting creatives as no longer valuable and targeting them for professional destruction. In an auto-immune disease, the carnage is pretty gory on the cellular level. Shreds of burst red blood cells, frayed neurons and optic nerves, inflamed and eroded joints.

In an automated society, we may not see corpses on the streets or in corporate offices, but the damage is as pervasive as it is acute. The creatives who choose not to allow an AI to get between them and their work and add “prompt engineer” to their CVs might be left behind financially as the generative AI sector keeps plowing ahead. Artists, writers, and other creators have had their work scraped by the algorithms without their knowledge or permission, copyright be damned, and they have certainly not been compensated. In a survey I’m running to gauge generative AI usage among marketers, early results show that part of the reason some marketers are using generative AI is that AI “lowers my content creator costs—I don’t have to spend so much on writers and designers.” Some of the respondents also appreciate the fact that these AI tools work 24x7 and don’t mind if the client gets grumpy or demanding. Yes, just take all that messy human relationship and interaction stuff right out of the equation. Who wants to work with another human, gross.

How does that make you feel, dear creative? You used to be “healthy tissue” in this body societal. Your art, your music, your words have held society together for millennia, but it’s all coming apart now. You have been reduced to a number—and the painful irony is that you are the wrong number. You are now a net loss on the balance sheet, while the pixels you generated over your lifetime are a net gain for generative AI companies, and likely the clients using their tools.

To be fair—generative AI is not a data vampire out to suck creatives’ blood (well maybe a little—artists, be sure to always include a picture of garlic in your online portfolios!) Artists are using these new tools with some stunning results. I personally have been entranced by a morphing animation loop one digital artist created. But here’s the thing: it’s their choice and intention, their art and their invention. It was neither their choice nor intention to have an invisible algorithm scour the arteries of the Internet and suck up the years of training, effort, imagination, and emotion that went into their work. It was neither their choice nor intention to have their worth devalued and diluted by the rising tide of content that now threatens to engulf the entire global web.

Amazon is clogged with books and other products of less-than-stellar quality. The Internet suffers from contentiac disease, its bowels ridden with filler blog posts, incessant ads, and keyword stuffers. Social media has built up life-threatening levels of narcissistic plaque; its blood pressure is through the roof. And YouTube? A case of monumentally botched Botoxic operation. Let’s not even go there.

To be still more fair, this didn’t happen yesterday, and AI is not the San Andreas fault line of the cataclysmic data-driven economic earthquake that has been ricocheting around the global economy. It’s just the tsunami triggered by the quake.

One human telling another human they are no longer valued—because they weren’t truly valued to begin with.

History teaches us we don’t need technology to be malicious to one another. How many wars have we fought. How many bodies eviscerated on the battlefield. How many psyches gaslighted, how many minds warped by psychological manipulation and torn apart by trauma. Somehow we always manage to effect harm—either on ourselves, on others, or other living things.

Artificial intelligence is only the latest twist in this long, winding road humanity is traveling. Not too long ago we had crypto. Before that, social media. Before that, smartphones—and pagers and faxes and CD players. Yes even the TV, the refrigerator, and the car were futuristic incredible things, once upon a time. We like shiny new things. But the shiny new things don’t necessarily like us back—not because they hate us or want to annihilate us (trust me, they really don’t), but because they are not alive to experience any sort of emotions. They’re our creations. Trouble is, we don’t always know how to use our own creations to our greatest benefit.

Whatever “intentions” or “impacts” we assign or attribute to these creations, are just tactics to deflect responsibility. You’ve no doubt seen the headlines screaming that ChatGPT is coming for our jobs. But it’s not ChatGPT, or any other AI. It’s the humans using this technology. That’s what Elizabeth was talking about in her Note—it’s the clients, the bosses, telling the writers they’re on notice. One human telling another human they are no longer valued—because they weren’t truly valued to begin with. Let’s be real: if a client really respected and valued your work as a writer, would they be quite so quick in replacing you with a half-baked language bot? If ChatGPT really were so much better than a human writer, how come we’re seeing media publications and companies starting to come out with AI rules and restrictions?

Point is, just because we now have the tools to do the work others have trained for years to do, and have dedicated their entire professional lives to, does not make all of us the same type of professional. I’m a writer, but I can’t sing. I’d love to be an opera singer in my next life, but in this one, I’ll let the rock bands get up on stage and cheer them on from the floor. Let the artists paint, the writers write, the singers sing, the programmers code, the bakers bake. Let us, all of us, enjoy our work.

The goodness of human things

So if your client or boss tells you to get up to speed on ChatGPT or else, thank them. Yes, THANK them—for letting you know up front that they do not value your work, your experience, your creativity and talent, and for giving you permission to work with others who do.

I remain the eternal believer in the goodness, integrity, and intelligence of humanity. No matter how much we mess up, we make up for it a thousand fold. After all…

Who is calling for more thoughtful deployment of AI and other technologies, to protect the privacy, security, and talent of individuals? Who is working to ensure greater recognition of women in AI?

Humans.

Who’s working through the night to crack the code of terminal illnesses? To find a viable alternative to plastics? To protect our forests and wildlife? To eradicate homelessness and child poverty?

Humans.

Who is tucking their children into bed at this very moment, having read two bedtime stories and dishes still to clean? Who is sending their hard-earned money to their family overseas so they can pay the rent and buy food? Who are all the unsung heroes who keep us safe and all of our systems running, that no one writes about?

It’s all us, humans.

We may be an auto-immune disease to ourselves, but we are also the cure.

Give a gift subscription

Delighted to announce my first guest post has been published on The Mindful Librarian, published by the lovely Kate Olson.

There’s a great storm brewing out there. You’ve heard it… the earth-shifting roar, the crack down the middle of the way we do life and work. Can you feel the rising tide of AI welling up and carrying all of our boats to the top of a monster wave? A wave none of us know how big it is, when it will crash, and which boats will crack under its force–and which will burst forward with irresistible energy. The wave has swept around the planet in a few short months, capturing 100 million+ people in its spell. Millions are enraptured, leveraging generative AI’s muscle-bound power in a dizzying array of applications and use cases, but too often doing so without thinking through implications, repercussions, costs, consequences.

Ain’t that the way we’ve always done things.

Except this time, those “repercussions” might be a little more serious.

Fiction writers, it’s time to roll up our sleeves, pull on our work boots, and do some serious critical thinking, lest we be swept out to the AI sea and lose our story bearings. So click out of those prompt engineering webinars, brew a strong cup of your favorite coffee or hot chocolate, and head on over to Kate’s Substack. See you there!

Edge case

/ˈedʒ ˌkeɪs/ 

noun

A problem or situation, especially in computer programming, that only happens at the highest or lowest end of a range of possible values or in extreme situations.1

Edge case (autonomous cars)

noun

In sectors like AI for autonomous cars, which operate in the physical realm, edge cases are all the things that can go wrong in the real world that you never generally do not expect.2 Things like children, squirrels, and soccer balls entering the road at the most inopportune times or the most inconvenient angles of entry.3

We’ve gotten quite friendly with computers over the years. Perhaps not quite as friendly as with our cars (I’ve named my car but not my laptop), but we cannot seem to let go of the notion that our smartphones and laptops and other devices are our friends, and continue to anthropomorphize the hardware and software we use. The current AI craze has deepened that tendency: we now use words like hallucinate, plagiarize, write, draw, think and create in relation to generative AI systems. These are words that in the not-too-distant past were exclusive to humans.

We also do the opposite: we refer to biological or living things in computer terms. We talk about uploading or downloading info into/from our minds, or say our mind glitched when we can’t remember some detail we wanted to share. I myself jokingly refer to my aging hardware when I can barely keep up with my pre-teen on a 9-mile bike ride. “As long as the software works,” I say, and my fellow moms laugh, knowingly.

As entertaining as it is to spin up metaphors and analogies to the way the human mind and computational machines mirror each other, it also carries a fundamental gap. It’s a gap that’s about as deep and wide as the Mariana Trench.

And that gap is lined with edge cases.

Let’s get edgy

Not to be confused with edge computing, edge cases are those instances that lie outside of the norm, at the edges of the expected and customary. The exceptions to a desired rule. Consider the case of the autonomous car, one of the most passionately debated examples of AI’s difficulty in mastering edge cases. The algorithms that guide the car’s movements must process a dizzying amount of visual and motion data—all in real time. As any human driver knows, you can travel the same road every day for twenty years, and the driving experience on that road will be different each and every time, largely due to things that we experience as a natural part of our physical world, such as the weather, the season, and the time of day; the condition of the road; traffic flows; pedestrians, cyclists, birds, animals, and insects; and of course, the other vehicles. Not to mention our own mental, emotional, or psychological state: we might be tired or energized; anxious or optimistic; calm or angry. These things are not specific to the road we drive. They are a part of our internal and external worlds. But they do define the experience of any given period of driving on any given road. 

Many of the external elements of the physical world are expected, and therefore fairly easy to program: the fact that a car drives on a road (as opposed to in a river, I guess). Other vehicles, moving or parked. Trees lining the sides of a street. Pedestrians, standing or walking. Traffic lights and stop signs. Daylight and nighttime. But here’s the kicker: what happens when you take any one of these expected external elements, and throw a twist into the equation?

These “expected elements” become edge cases.

The “pedestrian” might be a mother with a baby in a stroller and a young child holding her hand; the child sees a puppy on the other side of the crosswalk, lets go and starts running toward it. The mother is of course terrified, and runs after the child—with the stroller in tow.

The “tree” that’s supposed to stand along with other trees on the side of the street, has come crashing down (hello winter storm in California) and the road is impassable.

The “stop sign” that’s usually embedded on a post that’s stuck in the ground, is suddenly walking (no, it’s not a zombie stop sign, it’s held by a construction worker who’s helping reroute traffic).

The “other vehicle” has no driver or passenger and is driving itself in a circle in an intersection, while its owner is running after it, desperately trying to get back in (this has actually happened, in a small town in Connecticut some decades back—and likely plenty of other small towns across America).

For a human driver, it takes all of a second or two to take in the scene and react. Granted, this assumes they’re not distracted or under the influence. But the driver’s mind is able to recognize any given permutation of a pedestrian, tree, vehicle, or stop sign, and make quick decisions about how to maneuver their own car to avoid a collision or accident. (In the case of the driverless car running in circles, I imagine everyone at that intersection had a good laugh. Little did they know “driverless” would turn into “self-driving” in the not-too-distant future…) For our purposes here, whether or not the accident is materially avoided is secondary in importance to how the mind processes these “edge case” events.

To be fair, more than likely a self-driving AI system would pick up on a large tree blocking the road, and slow the car down. It should also “see” the mother chasing her toddler in the cross walk. And yet, there are instances of self-driving cars not seeing or recognizing pedestrians, running into various types of obstacles in their path small and large, and confusing red signs on buildings with stop signs. It’s this apparent inconsistency in edge case recognition that’s proving to be a thorn in autonomous AI engineering teams’ sides.

We can assume an autonomous AI system has been programmed with the same basic concept of driving that human drivers are taught: things like stopping at red lights and stop signs, driving on the right side of the road (in most countries), using the turn signal, following the speed limit, turning the headlights on at night, and so on. These scenarios form the core knowledge base of both the human driver and the AI. But the algorithm does not “see” the way we see—it processes visual and spatial data coming in from its sensors. It does not hear, feel, smell, or experience proprioception the way we do (yes, proprioception does come into play when you drive—in fact, experienced drivers have the ability to extend this sense of their body’s movement and position in space, to that of their vehicle). It does not feel anxious if it’s running late or impatient if a light stays red a little too long, and it’s not prone to road rage (thank heaven for that).

Most critically, the algorithm has neither the breadth and depth of the human mind, nor the gut instinct of an experienced driver. It can only do what it’s been programmed for and trained on, and eventually what it’s capable of learning on its own. Bottom line, the manner in which a human driver reacts to any given scenario differs from that in which an algorithm processes it. In some cases, the human will prove the better driver. In others, the algorithm—the car has sensors all around the body, for example, giving it a few extra pairs of eyes.

But even if the algorithm is able to handle some, or even many edge cases, the reality of our world is such that there will always be some scenario that is crazier, more unusual, or unforeseen, than the ones the algorithm is able to handle. This is not to say we’ll never have a sufficiently advanced autonomous car AI to hang up our car keys; but it does mean we should not labor under the illusion that self-driving cars will never make the wrong decision.

To illustrate the context a little better, here’s an edge case for you… if you’re in Iowa. Maybe not so much of an edge case in Alberta, Canada, where this photo was taken. :)

The reason I chose the autonomous AI sector to illustrate the concept of edge cases is because so many of us can relate. Raise your hand if you’ve survived at least one near-panic attack triggered by a squirrel that just couldn’t make up its little rodent mind which way to dart across the road: straight across; halfway in and stop; across and then back again to the original spot; wait till your wheels are about to crush it; or just sit there at the edge of the asphalt making you slow down for no reason. It’s one furry bundle of edge cases. Cute not cute.

It’s the real world, baby

The reality is that the world is a massive, planetary amalgamation of edge cases. If you like, you can think of it as one giant squirrel crossing the road (since the chickens are on sabbatical). What’s more, none of these edge cases are isolated from other potential edge cases. The squirrel might not have run into the road if it hadn’t been scared by an errant soccer ball. The small child might not have tried to run toward the puppy if there was no puppy, or if it had been a larger dog.

Except that these are not edge cases. Everything that happens, to us, near us, or even somewhere far away from us, is simply the real world, and it’s all deeply interconnected. Fundamentally of course, the squirrel is not an edge case. It’s a mammal acting as it has for its 36 millions years of existence. The bears in the photo above are not edge cases either. Nor is the uprooted tree or the handheld stop sign or the panicked mother in our discussion above.

The term “edge case” refers not to the actual animal or plant or thing, but to the frequency and (ir)regularity that its existence, activity, or occurrence represents. In other words, a squirrel crossing the road might represent an edge case in some scenarios, but a fairly routine occurrence in others. Think New York City vs Charleston, West Virginia. Or the bears in Iowa vs Alberta.

It’s there that the core of the matter lies. It’s one small associative step from recognizing the source or cause of an edge case as a real, material living being or thing, to linking the term edge case with that being or thing and conceiving of them as one and the same. For all AI intents and purposes, the squirrel, the stop sign, and the child effectively become the edge cases.

There is nothing strange or wrong about employing terms that aid in identifying and resolving mathematical or engineering problems and challenges. In fact, it’s necessary. Engineering, logic, and mathematics all need their own language to function and communicate, just as do law, philosophy, neuroscience, literature, art, music, and every other human endeavor. There are entire books about semantics and the role it plays in human society.

The danger we need to watch for is when the semantics we employ start to displace our sense of our own reality. In the industrialized world, and increasingly in the developing world, technology is so deeply interwoven into nearly all aspects of our lives that we have begun to normalize tech-related terms, metaphors, and analogies. We talk about minimizing screentime before bed as part of our “sleep hygiene.” As if it were a natural part of our nightly facial. We’ve got a whole new lexicon of social acronyms, inspired—necesitated?—by the size of smartphone keyboards and the speed-typing nature of online conversations. Some people actually pronounce the acronyms LOL and OMG in their IRL conversations. (BRB might be a stretch but I’m sure someone somewhere has tried.) Say “swipe left” and “dating” and pretty much everyone will know what you mean. Cute, but maybe not so cute.

We are being systematically programmed to think, react, and interact in more and more predictable ways, to smooth the marketing funnels, help take the edge off those pesky edge cases, and feed the ever-hungry algorithmic feedback loops. (Shoshana Zuboff, the author of The Age of Surveillance Capitalism, has a few thoughts on that—700 pages’ worth, and they’re worth their weight in gold). There’s a lot more to unpack there… stay tuned for an upcoming post.

Coming full circle

And so we circle back to the argument made in the beginning of this essay, which is that the human mind and the algorithm are not as similar as the AI companies and the media would love to have us think. Do I think AI is an extraordinary tool that can do wonders for humanity and the planet (assuming, again, it’s utilized with beneficial intentions)? Absolutely. Do I think the technologists are well on their way to creating a master AGI (artificial general intelligence) that will solve all our problems? No. There is a lot more to the human experience than the processing of information and the organization and completion of tasks.

If evolution had organized biological life into the neat little data points and packets the algorithms seem to prefer, we wouldn’t be here. Accidents, mutations, variations, anomalies… this is the wondrous creativity and infinite imagination of nature, of evolution, of life itself. To call a given event or occurrence “edge case” might be technically accurate, in the context of a larger set of events or occurrences, but it sucks the life out of it.

Think of any field, profession, or sector. Healthcare. Education. Finance. Agriculture. Sports. All the arts. Everything we do and experience is a rich, organic, dynamic system in constant flux, constant motion, constant evolution. Yes, there is always a “core” that defines the basic concept and experience of that field, but enveloping it and coursing through it are waves and currents and swirls of edge cases. It’s precisely the edge cases that help define the field, sector, or profession. This is what gives flavor to life. What spices up our experience and deepens our understanding of the world and our relationships with one another.

To the engineers and technologists, I ask, are we spending too much of our precious time and resources trying to parse every aspect of the human experience into bits and pixels? Artificial intelligence is an extraordinary tool, but as any tool, it is only as good as the purpose it is given. A wheel is only a decorative circle if it’s not employed in transportation, mathematics, or manufacturing. Shredded, soaked and pressed plant fibers aren’t much to look at until someone takes a brush or pen to them and writes a book (or prints some cash). And who doesn’t love some fine fermented grapes! AI can work wonders in finance, health, biotech, automation, manufacturing, and supply chain management—in areas where it serves as support for the human experience. When it is deployed to replace or co-opt the very bedrock of the human experience, formed by creativity, imagination, and ingenuity, it ceases to be just a tool and becomes an agent of cultural and societal decay.

The bull’s eye misses the point

We don’t need AI to control every aspect of our lives. It’s great and useful and valuable in specific sectors and use cases—what engineers call narrow AI—but the human experience is best left to, well, humans. Personally, I neither want nor need a chat bot to write my novels, stories, or posts like these for me. I will be the last human writer standing tall against the call for us storytellers to become “prompt engineers.” I also happen to enjoy the physical company of my fellow edge-case generating humans, and I know for a fact I’m far from the only one. If everything worked like clockwork and everyone had the personality of a cookie template, we would all get bored in a hot minute. It’s the edge cases that keep us on our toes, make life interesting, and give us things to talk about. How often have you come home bursting to tell your partner about the perfectly non-eventful drive home you just had? Now, that time you saw a squirrel darting backwards across the road…

So the question becomes, do we allow the algorithms to bleed us dry and render us data puppets, mechanically jostled from one reaction to another by over-engineered viral posts, or do we embrace the warm, churning chaos of unpredictability that makes us who we are, algorithms be damned? After all, chaos in the natural world is what drives beauty, non linearity, and innovation. How very edgy.