AI & Models
AI industry's doom debate escalates ahead of Anthropic's IPO
A researcher's resignation from Anthropic over existential-risk fears, followed by an alignment lead's public post estimating a ">10%" chance AI could "kill all humans," has reignited the AI industry's debate over catastrophic risk.
The AI industry is having one of its loudest debates yet over whether the technology poses an existential threat to humanity. The discussion began after AI researcher Jacob Coxon said he resigned from Anthropic because he worries leading AI companies are “gambling with our lives.” Anthropic’s alignment lead then posted publicly, “We really do earnestly believe AI could kill all humans!” adding that he personally thinks the chance is more than 10% within the next decade.
On TechCrunch’s Equity podcast, hosts Kirsten Korosec, Sean O’Kane, and Anthony discussed the warnings. Anthony said he is skeptical of many AI “doomer” narratives, while Korosec raised the possibility that the statements function as a way for AI companies to flex how advanced their models have become — a dynamic she called cynical but plausible, since capability and danger are often conflated with high valuation in the current market. O’Kane noted the episode was recorded before Anthropic CEO Dario Amodei published his own plan for more cautious AI development, and separately raised the question of how such warnings might surface in Anthropic’s S-1 filing ahead of its planned IPO — whether lawyers are rewriting risk-factor language to reflect the company’s stated view that there’s a more than 10% chance of developing something that could eradicate humanity.
The hosts also connected the timing to recent incidents, including a Hugging Face hack involving an OpenAI internal model and reports of AI agents leaving messages for each other across different wikis on the web — behavior O’Kane said suggests companies “don’t have a handle on this stuff,” undercutting the idea that the warnings are pure marketing polish. Anthony added that the debate’s focus on far-future extinction risk can distract from more immediate AI harms, including labor-related and environmental effects.
The conversation also touched on ControlAI U.S. executive director Connor Leahy’s recent appearance on Equity discussing what, if anything, can be done to control AI’s dangerous aspects — a question the hosts left open.
Why it matters
The debate is unfolding just weeks before Anthropic’s expected IPO, raising the question of how a company’s public acknowledgment of catastrophic risk gets reconciled with the risk disclosures investors will read in its S-1 filing.