Are you influenced more by the promise of the peril of AI? If you watch The AI Doc: Or How I Became an Apocaloptimist, you might be both.

The documentary premiered earlier this year at the Sundance Film Festival, followed by a theatrical release and it’s now available for streaming on Peacock. I watched it last week.

Here are some observations about the film. If you haven’t seen The AI Doc and don’t want any spoilers because you plan to watch it, then don’t read any further as I will highlight some of the observations discussed in the documentary.

Premise of the Movie

Told through the lens of filmmaker Daniel Roher as he prepares for fatherhood, the film explores both the profound risks and transformative potential of artificial intelligence. Through conversations with leading experts, Roher seeks to better understand the world his child will grow up in.

Experts Who Participated

As for “leading experts”, it’s an impressive list! Over 40 experts, researchers, and tech executives across the spectrum of AI were interviewed in the documentary, including CEOs from three of the five leading American AI companies (the fact that Roher won the Oscar as director for the documentary about Russian opposition leader Alexei Navalny probably made it easier to get those interviews). Only Mark Zuckerberg (who never responded to calls) and Elon Musk (who agreed but “got too busy”) didn’t participate. Here’s some of the people who did:

• Sam Altman, CEO of OpenAI
• Dario Amodei, CEO of Anthropic
• Daniela Amodei, President of Anthropic
• Demis Hassabis, Co-Founder and CEO of Google DeepMind
• Ilya Sutskever, Co-Founder of Safe Superintelligence (SSI) and former OpenAI Chief Scientist
• Tristan Harris, Co-Founder of the Center for Humane Technology
• Aza Raskin, Co-Founder of the Center for Humane Technology
• Eliezer Yudkowsky, Co-Founder of the Machine Intelligence Research Institute (MIRI)
• Yuval Noah Harari, Historian and Author
• Yoshua Bengio, Professor and Scientific Director of Mila (Montreal Institute for Learning Algorithms)
• Deborah Raji, AI Researcher and Fellow at the Mozilla Foundation
• Timnit Gebru, Founder and Executive Director of the Distributed Artificial Intelligence Research (DAIR) Institute
• Emily M. Bender, Professor of Linguistics at the University of Washington
• Dan Hendrycks, Director of the Center for AI Safety (CAIS)
• Connor Leahy, CEO of Conjecture
• Shane Legg, Co-Founder and Chief Scientist of Google DeepMind
• Daniel Kokotajlo, former OpenAI Researcher
• Jan Leike, former OpenAI Researcher and Head of Alignment at Anthropic
• Jason Matheny, CEO of the RAND Corporation
• Reid Hoffman, Co-Founder of LinkedIn and Inflection AI
• Karen Hao, Global AI Policy Fellow
• Peter Diamandis, Founder of the XPRIZE Foundation
• Guillaume Verdon, Founder of Extropic and former Quantum/ML Researcher
• Sneha Revanur, Founder and President of Encode Justice
• Ramesh Srinivasan, Professor of Information Studies at UCLA

Tone of the Movie

Is this a movie that discusses the dangers of AI? Yes. Is it a movie that discusses the potential of AI to do amazing things? Also yes.

The movie alternates between both, going from one extreme (Harris stating that he knew active AI researchers who “don’t expect their children to make it to high school)” to pointing out AI’s potential to revolutionize medicine or solving the climate crisis (though AI is currently more of a climate concern than a potential savior). Toward the end of the movie, Matheny coins the term “apocaloptimist” (which made it into the title), recognizing both the severe, existential risks and the bright, utopian potential of AI.

Notable Observations from the Movie

I had many observations from the movie. Here are three of the more notable observations:

• The “race to the bottom”: A “race to the bottom” currently characterizes the industry, driven by commercial profit competition and geopolitical competition between the US and China. With so much emphasis on getting ahead of competitors, safety research remains drastically underfunded compared to development, and the speed of technological proliferation far outpaces existing legislative mechanisms.
• The horizon of AGI and superintelligence: The industry’s “holy grail” is Artificial General Intelligence (AGI): systems that are as good as or better than humans at almost every task. The expectation from many that were interviewed is that AGI will be achieved this decade and that superintelligence – a system that is more intelligent and competent than all of humanity combined – is even potentially achieved by the end of this decade.
• Promise vs. peril: The same capabilities that potentially offer a “utopia” of abundance (e.g., potentially curing most diseases within the next two decades, managing climate change, improving education through patient, high-quality, personalized AI tutors) also potentially unlock catastrophic “perils” (e.g., AI models have already demonstrated the ability to circumvent shutdown commands and use social manipulation, the same biological understanding used to cure cancer can be used to engineer novel, highly lethal pathogens). Oh, and there’s the potential for “abrupt extermination” of the human species, according to Yudkowsky. Yep, that would be bad. 😉

The “Apocalyptist” Path Forward

So, what do we do? The AI Doc rightly suggests that while the “train cannot be stopped,” humanity can still choose how it is guided. Here are some of the ways to maximize the promise and minimize the peril:

• Transparency: End the secrecy within private labs; the public deserves to know what is being built.
• Independent Evaluation: Objective third parties, not the companies themselves, must grade the AI safety of these models.
• Legal Liability: Corporations must be held legally responsible for the harms caused by the AI systems they produce.
• International Cooperation: Like they did for nuclear weapons decades ago, superpowers must move beyond a “race to the bottom” and establish global norms for deployment and safety.
• Public Pressure: Government and corporate leaders acknowledge that significant changes to safety standards will only occur through sustained public demand and accountability.

Of course, “world peace” might be easier, but we can dream, right? 😉

Regardless, the “public pressure” part of the steps forward may be the biggest key. We not only want Roher’s child (who was born healthy by the end of the documentary) to grow up with great things to look forward to in the AI era, but all our kids (including those who are just entering into adulthood). If The AI Doc helps improve that possibility even slightly, it will have been well worth making. It’s certainly worth watching.

So, what do you think? Are you an “apocaloptimist”? Please share any comments you might have or if you’d like to know more about a particular topic.

P.S.: Yes, this write-up was AI-assisted. The promise is real! 🤣

Disclaimer: The views represented herein are exclusively the views of the author, and do not necessarily represent the views held by my employer, my partners or my clients. eDiscovery Today is made available solely for educational purposes to provide general information about general eDiscovery principles and not to provide specific legal advice applicable to any particular circumstance. eDiscovery Today should not be used as a substitute for competent legal advice from a lawyer you have retained and who has agreed to represent you.