The Most Interesting Thing in A.I.

Will AI destroy the world, or transform it into one of abundance? Across two books and several papers, philosopher Nick Bostrom has envisioned a range of AI futures. He joins Nicholas Thompson to discuss the ethics of how we treat AI, whether AI has sentience, and why he believes we should keep building, even at the risk of annihilation.  Produced in collaboration with PwC. (00:00) Introduction to Nick Bostrom and Superintelligence  (01:56) How AI development matched Bostrom's predictions  (04:48) Recursive self-improvement: Are we there yet?  (07:40) Physical limits of intelligence and computational ceilings  (09:40) Timeline predictions: Next year vs. next five years  (11:46) Embodied intelligence: Can AI replicate motor skills?  (14:32) Centralization vs. democratization of AI power  (16:52) The race dynamics: One leader vs  many competitors (19:57) AI alignment: Making systems behave as intended  (21:37) The gap between model power and our understanding  (23:25) "Optimal Timing for Superintelligence" paper explained  (28:14) Swift to harbor, slow to berth: When to pause AI development  (35:20) Moral status of digital minds and sentience  (41:23) Building trust with potentially misaligned AI  (44:11) Where to invest unlimited AI research funding  (47:07) Closing: Should we say please and thank you to AI? Learn more about your ad choices. Visit megaphone.fm/adchoices

Why does AI answer the way it does? Even as models cite their sources, the question of “why” remains one of the most confounding in the industry, with huge implications for users and builders alike. Mozilla CTO Raffi Krikorian says much of the answer lies in open-source AI— letting users look under the hood to see what’s happening. It’s a compelling idea, one that could also impact safety and alignment. But can it thrive? And what are the risks of ceding control? In a deep conversation with Atlantic CEO Nicholas Thompson, Raffi describes a world where technology is liberated from a handful of corporations, shares his hopes and fears for AI, and reflects on his recent car crash involving a self-driving Tesla. (00:00) Introduction: The mystery of AI decision-making and the need for transparency  (02:25) Twitter, Uber, and the DNC: Raffi’s career history (05:01) How centralization changed Twitter into X (08:30) Why seven companies shouldn't control AGI  (11:38) Mozilla's mission: Building an open AI ecosystem like Firefox did for the web  (14:21) Is it strange that Google funds Firefox? (16:40) The four layers of AI openness: Compute, data, models, and developer tooling (22:20) Data ethics and provenance: Creating markets for ethically-sourced training data  (26:44) What counts as “true” open-source AI? (32:04) The risks of open source AI: Balancing accessibility with safety concerns  (35:56) Should powerful AI be restricted like nuclear weapons?  (39:17) Raffi's Tesla crash and the danger of automated complacency  (46:56) Preserving humanity in the age of AI: Avoiding the "WALL-E" future Learn more about your ad choices. Visit megaphone.fm/adchoices

OpenAI’s Sam Altman sits for an interview with Nicholas Thompson, CEO of The Atlantic, to discuss AI’s trustworthiness, its dangers, and its impact on young people. Altman also discusses his company’s pledge to “stop competing and start assisting” rival projects that approach AGI, and why he thinks we’re not there yet. In a thorough and wide-ranging conversation, Altman opens up about where he thinks AI will go next and the mysteries that he still can’t solve. Recorded at OpenAI’s offices in San Francisco. (00:00) Introduction (02:33) Is our understanding of AI keeping pace with its growth in power? (05:10) Is chain-of-thought the key to trusting a model? (11:07) Open source AI, cybersecurity, and "infected" agents (16:26) Have we hit recursive self-improvement? (19:18) What does AI do on Sam Altman's computer? (21:15) Why hasn't AI made an impact in business yet? (24:04) Will AI make the wealth gap worse? (27:01) Why do young people hate AI?  (30:23) The challenges of AI sycophancy  (33:36) Do you regret making AI so human-like?  (36:53) Synthetic data and "mad cow disease"  (39:22) The future of publishing and media  (41:54) Do we need neurosymbolic AI?  (43:40) Will you cooperate with Anthropic if they get to AGI first?  (47:42) What is your advice to parents who are anxious for their kids' future?  (49:46) If you had infinite resources, what would you pursue?? Learn more about your ad choices. Visit megaphone.fm/adchoices