Nate Soares on How AI Really Could Kill Us All | Odd Lots
Community Score: 50% | 384 views | 4d
0 community ratings: null thumbs up, null thumbs down
Since the Hugging Face incident, concerns over AI doomsday scenarios has entered the mainstream. But of course, some people have been warning about this for years, long before LLMs or ChatGPT. In their 2025 book If Anyone Builds It, Everyone Dies, Nate Soares and Eliezer Yudkowsky spelled out their vision of humanity's inevitable demise once superintelligent AI is built. In this episode, we speak with Soares, who serves as president of the Machine Intelligence Research Institute, about why he believes it's impossible to "align" superintelligent AI, and how exactly the technology can and will go rogue as it gets more and more advanced. Chapters: 00:00:00 - Cold Open: The Path to Self-Sufficient AIs 00:00:42 - Podcast Introduction and AI Existential Risk 00:06:40 - The Origins of AI Safety Concerns 00:09:07 - Evolution of AI Alignment as a Field 00:10:41 - How Modern AI is "Grown, Not Crafted" 00:14:08 - The "Solid Gold Magicarp" Example 00:18:47 - The Fundamental Technical Cha
Tags: Joe Weisenthal, Massachusetts Institute of Technology, Tracy Alloway
More from Bloomberg Podcasts
- Follow the Money All the Way to Carney's Canada — Score: 50%
- Daybreak Weekend: US CPI, Europe Energy, RBA Decision | Bloomberg Daybreak: Europe Edition — Score: 50%
- Bloomberg Money: Credit Card Indicators and Golf's Changing Fanbase | Bloomberg Intelligence — Score: 50%
- We hear you! | Bloomberg Hot Pursuit! — Score: 50%
- Bloomberg Businessweek Weekend - August 7th, 2026 | Bloomberg Businessweek — Score: 50%
- Team Favorite: AI-Era Internet, Spaceport Investment, Bolivia’s Investability | Wall Street Week — Score: 50%