Calls Grow for Third-Party Lab to Verify AI Safety Claims

💯AI labs should not take public concern for granted. If the public is expected to be responsive to extraordinary capability or safety claims, we also need mechanisms for independent verification. https://twitter.com/bgurley/status/2080763671022616855

I largely agree with this point. 1) I thought this is what Anthropic was supposed to be; 2) claims like “open source is demonstrably safer” overstate the case and need to be removed from our discourse — just say this is a clear safety benefit of open source instead!… Show more

100% agree with @bgurley: we need a third party lab to evaluate these claims. https://twitter.com/bgurley/status/2080763671022616855

There's ample precedent in how to deal w/adversarial attacks on compute. The security apparatus in place for Windows, Mac, Linux, and a host of infrastructure is known, works, applies to OSS and proprietary products. It just requires companies to cooperate. Ball in ANT's court.… Show more