New AI battle: White House vs. Anthropic
Add Axios as your preferred source to
see more of our stories on Google.
Illustration: Aïda Amer/Axios
The White House and Anthropic, maker of the Claude chatbot, are in a war of words over AI regulation.
Why it matters: AI may be the century's most consequential technology, possibly even determining the geopolitical order, and rules are (or aren't) being written right now.
Catch up quick: Jack Clark, Anthropic's cofounder and policy head, on Monday shared a short essay titled "Technological Optimism and Appropriate Fear."
- It argues that too many people are pretending that AI cannot threaten humanity, and that we need to acknowledge a different reality before figuring out how to "tame it and live together."
- White House AI czar David Sacks responded by claiming that "Anthropic is running a sophisticated regulatory capture strategy based on fear-mongering."
Behind the scenes: The fight is as much about state-level regulations as it is federal ones.
- The White House supported a 10-year moratorium on state-level AI laws, proposed as part of the Big Beautiful Bill negotiations, arguing that 50 different rules in 50 different states would sow chaos and slow innovation.
- Anthropic called the moratorium "too blunt" and, after it failed to become law, endorsed a major piece of AI legislation in California.
Zoom out: Both sides support some sort of federal policy, although Sacks' driving philosophy so far has been to unwind federal safety work and "let them cook."
A big question for the White House is if Sacks is being hypocritical when accusing Anthropic of "regulatory capture" — given that Sacks and others on the White House AI policy team hail from monied tech interests in Silicon Valley.
A big question for Anthropic is if it's being hypocritical when it expresses "appropriate fear" while spending (and raising) billions for the sake of AI advancement.
The bottom line: Anthropic CEO Dario Amodei publicly supported Kamala Harris for president, and has been noticeably absent from White House tech events — ceding that ground to rivals like OpenAI.
- His company's regulatory efforts may help it recruit other safety-first researchers, but it won't help Anthropic gain traction in Trump's D.C.
