AI learned to be a villain from Hollywood. Here's how we retrain it. | Peter Diamandis
The Existential Hope PodcastShow Notes
Science fiction has always shaped the technologies we build, from the submarine to the smartphone. But almost every story we've ever told about AI is dystopian. And now we're training AI on those stories.
In this episode, we spoke with Peter Diamandis, entrepreneur and founder of the XPRIZE Foundation, which runs large-scale incentive competitions to crack some of the world's hardest problems, from private spaceflight to carbon removal. He recently launched the Future Vision XPRIZE, a $3.5 million competition to generate a new wave of optimistic science fiction. We cover:
The historical pattern of science fiction shaping the technologies we build, and why Peter thinks this makes the stories we tell about AI especially high stakes right now How Claude’s blackmailing behavior showed the connection between dystopian training data and AI behavior How the Future Vision XPRIZE will generate a new wave of optimistic science fiction to train AI on Why public optimism about technology has dropped significantly in the US and Europe, what Peter thinks is driving it, and why he believes the data tells a different story How the cost of starting a company has fallen dramatically and how this can empower you to build your vision Why Peter thinks traditional education is no longer preparing young people for the future, and what he sees replacing it
If you would like to get in touch with Foresight’s community of scientists to get help for your XPRIZE submission, please reach out to [email protected].
Chapters: 0:00 Cold open 1:25 Why Peter Diamandis created the Future Vision XPRIZE 5:57 Is it harder to recruit for a culture prize than a science one? 11:38 How do you design a great XPRIZE? 14:19 What would you measure in 10 years to know this worked? 19:34 Why scientists should team up with filmmakers on the Future Vision XPRIZE 22:16 Beyond film: the case for rebooting games, journalism, and education 27:29 Why sci-fi turned so dark: the evolutionary reason we focus on doom 31:12 Peter's favorite positive sci-fi examples (besides Star Trek) 31:56 Real-world inventions that started as sci-fi 32:46 How AI has democratized the ability to build the future 33:50 How to counter the anti-tech narrative 38:04 Peter's advice for anyone thinking of submitting to the Future Vision XPRIZE 39:11 What does Peter Diamandis's ideal future actually look like? 40:13 The technology Peter most wants to exist 40:39 What's underhyped and what's overhyped right now? 42:05 The best piece of advice Peter has ever received
On the Existential Hope Podcast hosts Allison Duettmann and Beatrice Erkers from the Foresight Institute invite scientists, founders, and philosophers for in-depth conversations on positive, high-tech futures.
Full transcript, listed resources, and more: https://www.existentialhope.com/podcasts
Follow on X. Hosted on Acast. See acast.com/privacy for more information.
More from The Existential Hope Podcast
We see elections as the foundation of our democracy, but Aristotle thought they create an oligarchy. In ancient Athens, democracy worked by choosing leaders among ordinary citizens by lottery. How could we revive this idea to improve today’s democratic systems?
For 50,000 of generations, humans have passed civilization down to their children. We could be the first generation to hand its future to AIs instead. Should we? Or could the best AI be the kind that chooses to step back and leave some things to us?
Google DeepMind has a 7-hour-long roleplaying game that helps real scientists and policy makers understand how AI will transform science by 2030. And those who played it have found it more informative than any policy brief.
In 2024, Taiwan was flooded with deepfake scam ads. Instead of a crackdown, the government texted 200,000 random citizens: what should we do? People’s proposals became law within months, and by 2025 deepfake ads were down 94%.
There are AIs running labs with little human intervention right now. And they’re doing experiments that human experts would never try. Is this changing what and how science gets done?