Anthropic breaks with rivals over open-weights industry letter

For a company that spends a lot of energy positioning itself as the grown-up in the frontier-AI room, Anthropic is having a bruising month. The Wall Street Journal reports that resentment inside Silicon Valley toward Anthropic has hardened, with the flashpoint being an industry letter titled 'Open Weights and American AI Leadership' that more than twenty tech companies, including Meta, Nvidia, OpenAI, Google, Microsoft and SpaceX, have signed. Anthropic did not.

David Sacks, the White House AI and crypto adviser, has been the loudest external voice. He posted on X that 'the entire tech industry (save for Anthropic) has come out in favor of open source AI' and warned that 'they won't stop until they kneecap open source.' Separately he accused the company of running 'a sophisticated regulatory capture strategy based on fearmongering.' CEO Dario Amodei answered on July 27 with a lengthy statement saying 'Anthropic has never advocated for a ban on open-weights models,' describing models without dangerous capabilities as 'a public good.' Reid Hoffman, whose Greylock is an Anthropic investor, has publicly defended the company as 'one of the good guys.'

Why this matters beyond the personalities is that Anthropic has spent two years building an identity as the safety-first frontier lab, and that identity is now being reframed by rivals and by the White House as commercial self-interest dressed up as policy concern. The trigger, per the reporting, was the shock debut of the Chinese open-weight model Kimi K3, which reset the political conversation about whether restricting open weights protects American leadership or hands it away. Google and OpenAI reportedly signed the letter only after initially abstaining, which tells you where the wind is blowing.

The honest caveat is that the reporting is heavy on Silicon Valley sentiment and lighter on measurable commercial damage. It does not enumerate lost enterprise contracts, defected partners, or a specific hit to Claude adoption. Anthropic's stated position, that it wants tighter chip controls, action against industrial-scale distillation, and mandatory testing for sufficiently capable models rather than a ban, is arguably not as far from the letter's signatories as the fight makes it look.

What's worth watching is whether the 'regulatory capture' framing sticks. If it does, every future Anthropic statement about safety will be read through a competitive lens, which is a much harder environment for a company whose whole brand rests on being trusted to make judgment calls about capability.