The AI Evaluator Forum released a letter outlining minimum conditions it believes labs such as OpenAI and Anthropic should meet to work effectively with third-party risk evaluators. The letter followed commitments by OpenAI CEO Sam Altman and Anthropic CEO Dario Amodei to bring in outside evaluators. Computer science pioneers Geoffrey Hinton and Stuart Russell were among the dozens of signatories. The forum includes several watchdog groups focused on AI risk.
Related Articles
Don't miss out on breaking stories and in-depth articles.