AI will only become more polarizing from here

The battle lines have been drawn. You're either for AI or against it. The enemy of your enemy is now your friend. That makes for strange bedfellows.

When Barack Obama is using his platform to raise awareness of the disruptive power of AI, you know it’s gone mainstream. MSNBC has also featured a “Morning Joe” episode talking about job disruption from AI. I’ve noticed that AI reactions tend to cluster into a few camps.

The skeptics tend to focus on numerical evidence, or a lack thereof. They cite “AI hasn’t dislocated enough jobs to show up in data yet.” However, any scientifically literate commentator will point out “absence of evidence is not evidence of absence.” Furthermore, just because AI dislocation is not yet statistically significant does not mean it is not happening, nor that it won’t ramp up.

Another group are the rapidly anti-AI crowds. Some are artists, others are “liberals” (which I use scare quotes because I’m a progressive but pro-AI since AI has the potential to equalize everything from education to healthcare), and yet others are Doomers—my arch nemeses, those people who are convinced, without evidence, that AI will surely kill everyone.

I’ve been accused by various groups of being right-wing simply for being pro-AI and not being “progressive enough” when I say that we should celebrate AI nuking the wage-slavery social contract and creating a more prosperous future via Post-Labor Economics as a way of getting there. I guess you’re not a “real progressive” unless you want to skin billionaires alive and roast them on the fires of burning data centers. Or something like that.

Keep in mind, I’ve also been accused of being a rapid “liberal marxist” since I advocate for decentralized and collective ownership models. It’s kinda ironic, since people can’t really seem to put me into a single bucket. Likewise, on the topic of AI itself, I don’t fit neatly into any single bucket. On the one hand, I am working to protect the economic agency of everyone in the transition we’re facing. I see the writing on the wall—the march of technological progress has only ever been held back by backwater nations.

Far right think tanks see this chart and declare it as “Marxist, Feminist, liberal nonsense” and meanwhile far leftists see it and say I’m a MAGA supporter because I believe in money. I kid you not, these are reactions I’ve gotten.

AI is coming, like it or not.

Many skeptics, deniers, and haters seem to think that just whining and screaming “shut it all down” or “nothing to see here” makes it true. This sort of “high conviction, high volume” rhetorical tactic is par for the course lately.

Folks like Shawn K, who reached out to me about his blog (linked below) really matter to me. Yes, I might sound flippant or dismissive when I write about the state of AI discourse, but I am not dismissive of the human element. What I really care about is the actual suffering that people are enduring—the depression, anxiety, and uncertainty. The reason I’m cynical about the state of discourse is because, from my vantage point, much of it is overly tribal and myopic.

On the one hand, we have clusters of people claiming “nothing to see here” and burying their heads in the sand. Certainly, skepticism serves a social purpose. However, it comes off as flat out denialism when real people are losing their livelihoods. Doomers and Dreamers are cut from the same cloth, ironically enough. Both of these groups trust their emotions, intuitions, and tribal vibes more than scientific evidence, consensus, facts, or numbers. They discard the engineering reality and systemic complexity in the real world and focus on “if I can imagine it, it must be real.”

I often need to clarify my techno-optimism. In the grand arc of history, technology has generally made life better for people. It has created new problems (like porn addiction) but it has solved many other problems like tuberculosis and hunger. Arguably, if social media algorithmic outrage is the cost of progress, I would prefer to live in the current time.

The Great Displacement Is Already Well Underway

Earlier, I said I saw this coming, but I didn’t realize how bad it would become.

When I first got into AI, we had SKLEARN and not much else. Then GPT-2 and GPT-3 came out. The arguments we had then were over topics like synthetic data and what we now call RAG (retrieval augmented generation). Back in 2021, not even the Doomer movement was all that serious. Their doomsday prophecies were still entirely hypothetical, something that would happen far into the future, and so their flawed armchair “logic” was tolerable.

Fast forward a few years, and now the stakes are higher for everyone. Yudkowsky has a book coming out entitled “If Anyone Builds It, Everyone Dies” And this is not hyperbole. He literally believes that despite his lack of education or technical experience. He’s convinced himself, and plenty of others, with his self-referential treatises on “thinking” that you can just logic your way to understanding how AI will work. No experimental evidence required, scientific consensus, or even peer review!

Strange times make for strange bedfellows.

By and large, the more energetic arguments break down across a single dichotomy: for and against AI.

As a lifelong Trekkie, I am of course pro-AI, but that means my allies are evolving. I don’t formally align myself with E/ACC but we are aligned against Doomers. Likewise, I don’t fully endorse the recent “China Hawk” phenomenon, but I know enough about history, economics, and geopolitics to understand that rivalry between the US and China is both inevitable, and likely good, in the long run.

These people have spent so long in their own echo chambers that they genuinely believe this.

The anti-AI crowd, likewise, is finding itself with strange clusters of allies; artists, “liberals”, and some ordinary workers are joining forces with the Doomers.

As with many of my takes, I tolerate a tremendous amount of nuance and ambiguity. Nothing is ever black and white or clearcut. Many folks like to paint solutions with “big crayon” thinking—just shut it all down, just use Bitcoin, and many other “just do X” kind of solutions. Reality has never worked that way, and never will.

I feel like I alone, along with a small cohort, look at all sides and say “wow, this is a complicated mess, and it’s not going to be easy to solve, let’s actually immerse ourselves with this complexity” rather than just rejecting it in favor of black-and-white thinking.

As valence goes up, tempers flair, and people will care more and more. The more energy there is in a domain, the more heated it becomes. It’s all thermodynamics, and the pressure cooker has only just now switched on.