Guest post: AI is not a democratizing force

It is, in fact, anti-democratic.

Image courtesy of PauseAI via the platform formerly known as Twitter

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.

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.