Anthropic Governance Tested by Departing Researcher Amid $965 Billion Valuation

Anthropic, an AI company backed by Amazon and Alphabet, lost a researcher named Jacob Coxon before any of his equity vested. The departure is seen as a governance and valuation signal rather than an operating crisis.

Coxon was only at the company for four months but had worked on pretraining at OpenAI and Anthropic. He alleged that frontier labs are racing toward self-improving systems without enough coordination on safety, raising concerns about what could happen under competitive pressure.

Anthropic has consistently acknowledged both the benefits and risks of advanced AI and supports a lawful, verifiable mechanism for labs to coordinate the release of powerful models. The company's last financing valued it at $965 billion post-money in May, with an annualized revenue run rate of $47 billion.

Amazon's sensitivity is much larger than its original cash investment suggests, with a combined carrying value of $190.4 billion in Anthropic nonvoting preferred stock and convertible notes. Alphabet also has a stake in Anthropic but the number is opaque, with its commercial dependency on the company being tens of billions of dollars.