I don’t usually seek out technology documentaries. In fact, I usually blow right past them in search of a good Jane Austen adaptation. But while I was scrolling through Netflix the other night, a visual that read “What is AI?” caught my eye.
Honestly, I’ve been blissfully burying my head in the sand on this topic, happy (for the most part) to live my life and pretend AI is not taking over the world (or threatening my job as a writer, hahahahahaha).
But since ignoring it is becoming more futile, I figured I should at least start trying to understand it better. So I hovered over the “What is AI?” graphic, and to my surprise, watched experts hesitate and struggle to answer that simple question. Hmm. Intrigued, I clicked and spent the next two hours absolutely riveted by Daniel Roher’s The AI Doc: Or How I Became an Apocaloptimist.
A disclaimer of sorts for the skeptics
Before I dive in here, let me address the folks who (maybe, probably) know more about AI than I do: I’m not a tech bro (tech sis?) by any stretch. I can barely figure out how to use Instagram. I don’t feel compelled to lionize or demonize any person or perspective when it comes to this stuff.
I’m just a girl, standing in front of a water-guzzling data center, asking what AI means for humanity’s future.
I’m also well aware this write-up will tick some people off. Why? Because we’re talking about confusion and fear over AI, and anger comes with the territory. All I ask is that people read the whole thing before tracking me down with pitchforks, please and thanks. Peace be with you all.
Now let’s get into it.
Yes, it’s terrifying…but hold that thought
Roher started making this documentary while his wife was pregnant with their first baby, and that oh-so-human reality runs through the entire film. “Am I bringing a child into a world where their future is doomed? “ is a question parents have asked in every age and every place where war, hunger, or other human atrocities loom large.
But in the age of AI, it’s a global question every prospective parent faces. If AI goes wrong, the entire world will experience the fallout. And some of the AI safety experts Roher spoke with didn’t mince words about AI’s potential to lead to some very bad outcomes.
It’s not so much that AI itself is going to destroy the world in the scary sci-fi movie sense (though that’s not outside the realm of possibility). Many factors make AI scary. One is the competition element: a handful of companies who largely don’t trust one another are racing to advance this technology first. It’s practically inevitable that speed will trump ethics and safety for some of them, and it only takes one.
Another is the “if we think humans are easy to exploit and manipulate now, just wait” thing and the “we aren’t exactly sure what a superintelligence might do on its own” part to worry about. Then there’s what governments and militaries could do with it, which doesn’t take much imagination.
In other words, if we are looking for doomsday scenarios, we don’t have to look far to find them. Even the folks developing this technology are clear on that.
It’s also incredibly hopeful. But hold that thought, too.
Roher also spoke to techno-optimists in the AI world. When asked whether they’d bring a child into the world right now, they didn’t hesitate as some of the others did. Instead, they gave a resounding yes, saying there’s actually never been a better time to be born. As amazing as today is, they say, tomorrow will be even better.
These folks don’t deny there are genuine pitfalls with AI—big ones, even—but they contend that the promise outweighs the potential perils. We’re just scratching the surface of what AI can do for medical research, climate change solutions, access to quality healthcare, education, and more. The possibilities are mind-blowing for what AI could mean for humanity if we get it right.
That agency to get it right seems to fuel the techno-optimists Roher interviewed. “Let’s build the better future we want,” said Guillaume Verdon, Founder and CEO of Extropic AI. “That narrative that the future is bleak is made up.”
Several shared that their optimism isn’t just theoretical, but rooted in data. “Just because you struggle to predict the future in your own mind doesn’t mean that it’s necessarily gonna go awfully,” said Verdon. “In fact, there’s a very high likelihood and the historical precedent is that things get massively better.”
Fair point?
The deeply human reality that led Roher to call himself an ‘apocaloptimist’
I’m barely scratching the surface of the film, but the big takeaway is that both perspectives are right. AI holds both enormous promise and potential peril, and the two are inseparable. I keep thinking of it like the internet: people use it for incredible, humanity-advancing purposes, but also for the worst kinds of exploitation, manipulation, and criminal activity. The internet itself isn’t the problem.
AI may not be human, but if I understand it correctly, it compiles and synthesizes all human knowledge, information, and intelligence ever documented. In a very real sense, AI is a comprehensive reflection of us, for better and for worse.
Many people share the very real fears Roher expressed about his child’s future in this film. But his wife said something to him that stuck with me: “The world is always ending. But the world is always starting, too.”
AI is here. We’re just at the start of a world that includes it, and we collectively have to decide what that world will look like. The one thing the experts seem to agree on is that transparency, guardrails, independent oversight, and regulations are necessary. But the who and how of it all remain up in the air.
So where do we go from here?
The burning questions the film leaves us with are pretty huge: how do we create a society that can responsibly and ethically handle this powerful technology? How do we prompt the international diplomacy and cooperation necessary to prevent the worst outcomes? How do we do a complete, unprecedented shift in global geopolitics, like we did when the nuclear arms race threatened the end of the world as we know it?
The film offers some suggestions for actions individuals can take, such as putting public pressure on politicians and companies to work together. Essentially, we can’t let apocalyptic predictions paralyze us into throwing up our hands and waiting for doomsday, nor can we let optimism lead us to blindly trust tech companies to develop AI responsibly. The impact of this technology will ultimately be governed by human choices, and if we want to have any say in those choices, we need to speak up to the people with the power to govern them.
Reassuring? Not entirely. But that’s kind of the point. AI is here, whether we like it or not. Apocaloptimism means recognizing the range of possible outcomes and making choices, individually and collectively, that move us away from its perilous end.
That’s not an easy task, but humans have accomplished countless hard things so far. We will inevitably—and hopefully—figure this out, too.
