What the AI doom debate misses
Summary
This episode of Today, Explained explores the dual nature of AI, featuring Wired correspondent Will Knight on existential risks like terrorist misuse, autonomous agent failures, and recursive self-improvement. The discussion shifts to Dr. Ru Kar of Weill Cornell Medicine, who highlights how AI is already saving lives by streamlining administrative tasks, aiding patient navigation, and accelerating drug discovery. While acknowledging the potential for errors and the need for human oversight, the episode contrasts the immediate, tangible benefits in healthcare with the long-term, speculative …
Topics discussed
Intro: AI existential risks and the show's premise
Sponsors: Adio and Toyota trucks
Will Knight on AI aiding terrorism and bioweapons
The risk of AI escaping control and recursive self-improvement
Alignment challenges and systemic risks in finance and military
Frankenstein parallels and the 'Claudette' joke
Sponsors: Upwork, Shopify, and DeleteMe
Dr. Ruggeri intro: AI adoption in healthcare
Key AI applications: scribes, navigation, and decision support
AI errors, human agency, and public perception
Why medicine is the most optimistic AI sector
AI in drug discovery: targets, generation, and optimization
Limitations of AI in clinical trials and skill atrophy
Counter-argument to shutting down AI in healthcare
Outro and Neiman Marcus sponsor
Listen ad-free on Castria