In 1964, science fiction author Arthur C. Clarke predicted that computer systems would overtake human evolution.“Current-day digital brains are full morons, however this won’t be true in one other era,” he advised the BBC. “They’ll begin to assume, and finally, they may fully out-think their makers.”
Daniel Roher opens his new documentary The AI Doc: Or How I Turned an Apocaloptimist (2026) with this cheerful prophecy. And within the hundred-some minutes that comply with, he tries to make sense of a expertise that, by his personal admission, he doesn’t perceive — and a world that’s quickly being modified by it. Explaining that he conceives of AI as a “magic field floating in house,” he enlists the assistance of consultants to supply him with a crash course in what, precisely, AI is.
Roher’s actual concern, nonetheless, isn’t a lot concerning the workings of AI — although a few of his topics do try to elucidate them for him — however whether or not it’d displace us, as Clarke’s prediction suggests it is going to.
Whereas making the movie, Roher learns that his spouse Caroline is pregnant with their first youngster. He tracks his spouse’s being pregnant and the start of his son in parallel with the arrival of AI. It’s a wise alternative that builds on a worry all dad and mom share: What kind of world are we making for our kids? And behind that query is one other, vibrating in anxious silence: What occurs after our offspring exchange us? This twinned existential angst drives his efforts to listen to from the doomers, the techno-optimists, and the in-between “apocaloptimists” whose ranks he finally joins.
The AI Doc, as its sweeping title suggests, needs to form and lead the narrative round AI. It’s definitely arrange to do this — Roher is recent off an Oscar win for his documentary Navalny, and the movie opened in practically 800 theaters, which counts as wide-release for a nonfiction title. The ultimate product is indicative of the ways in which public attitudes round AI are in large flux. Roher hopes to succeed in folks of my grandmother’s era who conflate AI with smartphones and spellcheck, in addition to individuals who don’t appear to care whether or not a video was AI-generated.
However I feel that this documentary has come too late to steer the dialog, one thing the movie itself acknowledges. For all its transformative potential, AI isn’t really distinctive amongst rising applied sciences but — it has not been cataclysmic or ushered in a golden age of prosperity — however Roher and lots of of these he interviews are likely to deal with it as a radical break with all that has come earlier than. Consequently, they have a tendency to fixate on the binary extremes of doom or salvation. It’s an strategy that reinforces our personal helplessness within the face of AI-driven change, whereas additionally muddying our understanding of what we’d but have the ability to do as we search to adapt, mitigate hurt, and form the world that AI might in any other case really begin remaking.
Roher, considering his youngster’s future, opts to listen to the dangerous information first. Tristan Harris, the cofounder of the Heart for Humane Expertise, doesn’t mince phrases: “I do know individuals who work on AI threat who don’t count on their youngsters to make it to highschool.”
Lots of the movie’s different interviewees are equally gloomy. Geoffrey Hinton, the “godfather of AI,” for instance, argues that as AI turns into smarter, it is going to turn into higher at manipulating humanity. However nobody is extra pessimistic than Eliezer Yudkowsky, the well-known AI doomer and co-author of the controversial e book If Anybody Builds It, Everybody Dies. Because the title suggests, Yudkowsky believes that superintelligent AI would wipe out humanity — a place that he stands by and lays out for Roher.
Turning his again on these storm clouds — and taking the recommendation of his spouse, Caroline, who tells him that he wants to search out hope for the long run — Roher tunes into the refrain of AI optimists. They inform him, variously, that there are extra potential advantages than downsides to AI; that expertise has made the world higher in each manner; that this would be the software that helps us clear up all our best issues. To not point out: AI will deliver the most effective well being care on the planet to the poorest folks on Earth, lengthen our healthspan by a long time, and allow us to dwell in a postscarcity utopia freed from drudgery. Oh, and: We’ll turn into an interplanetary species, all due to AI.
These guarantees initially reassure Roher, maybe as a result of he appears simply led by whomever he’s spoken to most not too long ago. It’s Harris who finally convinces him that we are able to’t separate the promise of AI from the peril it presents. The conclusions that outcome can be apparent to anybody who’s considered these points for greater than a second or two: If AI automates work, for instance, how will folks make a dwelling?
It doesn’t assist that lots of the most invested gamers mirror on these questions superficially, if in any respect. OpenAI CEO Sam Altman tells Roher that he’s anxious about how authoritarian governments will use AI — a declare that’s adopted within the movie by a lower to photographs of Altman posing with authoritarian leaders. Different tech CEOs fall again on PR pleasantries in response to the filmmaker’s questions, and Roher too usually goes simple on them, by no means diving deeper once they admit that even they aren’t assured that all the things will go nicely. That these are the leaders of AI firms racing in opposition to one another to make the expertise increasingly superior does little to encourage confidence.
(A few of the techno-pessimistic folks interviewed for the documentary have expressed their sturdy displeasure with the ultimate outcome.)
“Why can’t we simply cease?” Roher asks these tech CEOs. He’s advised {that a} moratorium is a pipe dream: Many teams world wide are constructing superior AI, all with totally different motivations. Laws lags far behind the speed of technological progress. Even when we might go legal guidelines within the US and EU that might cease or sluggish issues down, says Anthropic CEO Dario Amodei, we’d should persuade the Chinese language authorities to comply with go well with.
If we don’t create it, the considering goes, our enemies will. It’s finest to get forward of them.
That is, after all, the logic of nuclear deterrence: If we don’t mitigate the chance of ending the world by mutually assured destruction, there’s nothing stopping another person from urgent the button first.
An apocalypse in each era
The atomic comparability is apt, if solely as a result of Roher sees the stakes in equally stark phrases. “Will my son dwell in a utopia, or will we go extinct in 10 years?” he wonders aloud. It’s a query that’s central to the movie. However he by no means actually sits with the extra doubtless situation that AI will neither result in human extinction nor finish all illness and drudgery. Each era faces the specter of its personal annihilation — and but the ends of days hold accumulating, irrespective of how shut the doomsday clock will get to apocalypse.
The purpose, then, isn’t that AI gained’t be dangerous for us, however that by framing the query in strictly utopian or dystopian phrases, we miss the messy actuality that lies between hell on earth and heaven within the stars. Though The AI Doc tries to chart an “apocaloptimist” course between two extremes, it doesn’t grasp the actual stakes. AI doesn’t actually create new dangers as such — it’s a power multiplier for current ones like the specter of nuclear warfare and the event and use of organic weapons. The chief existential dangers of AI are human-made and human-driven. And which means, as Caroline says within the movie’s ending narration, “We get to determine how this goes.” She’s proper, however her husband by no means appears to grasp how she’s proper.
Like too many Huge Problem Documentaries, Roher’s movie is heavy on issues and light-weight on options. It does provide some, calling for worldwide cooperation, transparency, authorized liabilities for firms if one thing goes mistaken, testing earlier than launch, and adaptive guidelines to match the velocity of progress. However simply as it is a strictly introductory course in AI — one that may most likely irritate those that’ve already moved on to AI 102 — these suggestions are solely a place to begin. For Roher, they provide motive to be hopeful. For the remainder of us, they’re just the start of a chance to meaningfully steer the course of our future.
You’ve learn 1 article within the final month
Right here at Vox, we’re unwavering in our dedication to masking the problems that matter most to you — threats to democracy, immigration, reproductive rights, the surroundings, and the rising polarization throughout this nation.
Our mission is to supply clear, accessible journalism that empowers you to remain knowledgeable and engaged in shaping our world. By turning into a Vox Member, you immediately strengthen our skill to ship in-depth, unbiased reporting that drives significant change.
We depend on readers such as you — be a part of us.
Swati Sharma
Vox Editor-in-Chief
