Anthropic Seeks Outside Input on Claude's Values, Keeps Final Say
TL;DR
- Anthropic brought law professors and democracy experts to its campus to debate who should ultimately set Claude's values.
- Reporting makes clear that Anthropic retains final decision-making authority over how outside feedback translates into practice.
- NBC traces the broader exercise to an earlier Anthropic conference held in London in May.
Anthropic has invited law professors and democracy experts to its campus to debate who should ultimately decide the values guiding Claude, while retaining final authority over how that feedback is used, NBC News reported.
The piece, written by Jared Perlo, situates the exercise against what it calls leading AI companies 'grappling with knotty questions about the values that guide their products' behavior,' even as they 'insist their technologies could usher in a new industrial revolution.' The convening follows an earlier Anthropic conference in London in May. The headline frames the unresolved governance tension bluntly: Anthropic wants outside input, but gets the final say.
This is our third Anthropic item on the tracker today, alongside the company's parallel consultation with law firms on AI-agent negotiation norms and Jack Clark's IMF Bangkok remarks on AI and GDP, a visible pattern of soliciting outside credentialed input while reserving the call for itself.
Originally reported by nbcnews.com
Read the original article →Original headline: Anthropic Convenes Law Professors and Democracy Experts to Weigh Who Decides Claude's Values