Anthropic Researcher Sounds Alarm 9/9/26
Summary
Anthropic researcher Jacob Coxon resigned after warning that the AI industry is racing toward self-improving superintelligence, effectively "gambling with our lives." He and other alignment experts argue that recursive self-improvement poses an existential risk, with one researcher estimating a greater than 10% chance of human extinction within the next decade. The episode discusses the tension between these internal safety concerns and Anthropic's ongoing IPO process, while noting that such warnings have been raised for years by figures like Elon Musk. Host Kelly and reporter Kate Rooney d…
Topics discussed
Edward Jones: Financial planning and the meaning of wealth
Schwab: Thinkorswim platform and Ameritrade integration
AI Safety: Anthropic researcher steps down over concerns
Warning: Race for superintelligence and existential risk
Internal dissent: Other researchers echo extinction fears
Technical risks: Recursive self-improvement and alignment
Business impact: Anthropic's IPO and SEC risk factors
Public perception: Boy who cried wolf and data center backlash
Historical context: Musk, OpenAI, and P(Doom) debates
Industry criticism: Dario Amodei and pessimism accusations
Regulatory outlook: Government response and open questions
Energy Transfer: Oil and gas in everyday products
Listen ad-free on Castria