Anthropic Researcher Quits Citing >10% AI Extinction Risk; $183B Valuation at Stake

The Quick Take

• An Anthropic safety researcher assigned greater than 10% probability to AI causing human extinction, the BBC reported.
• A second researcher resigned from Anthropic citing the company's race toward self-improving superintelligence.
• Anthropic's $183 billion valuation faces scrutiny as internal safety dissent raises questions about regulatory exposure.

Anthropic, valued at $183 billion (£135 billion), faces a rare moment of internal safety dissent: one researcher publicly assigned greater than 10% probability to AI systems causing human extinction, while a colleague resigned citing concerns about the company racing toward self-improving superintelligence alongside rival OpenAI. The BBC coverage — tier-1 amplification — brings these warnings into mainstream financial discourse for the first time since AI valuations reached current levels.

The market implications are twofold. First, regulatory risk: if credible insiders are publicly assigning double-digit extinction probabilities, regulators in the EU, UK, and US gain a powerful justification for imposing capability restrictions — the kind that could limit Anthropic's commercial product roadmap. Second, competitive dynamics: the researcher's accusation that both Anthropic and OpenAI are "gambling with lives" creates mutual reputational risk that third players (Google DeepMind, Meta AI, Chinese labs) could exploit.

The key watch is whether other AI company employees publicly echo these safety concerns — a coordinated signal would accelerate the regulatory response timeline. For Anthropic specifically, monitor its next funding round valuation (whether the $183B mark holds under this narrative) and any board-level response to the safety dissent. UK AI regulators have a mandate to act swiftly on credible safety reports, making this the most immediate geographic risk.

Synthesized from 2 sources.