Anthropic has selected Accenture as its first embedded evaluator, partnering with Accenture’s AI business Faculty to evaluate and red-team models, conduct alignment assessments and test safeguards. The non-exclusive collaboration supports CEO Dario Amodei’s Sept. 12 proposal for independent evaluators with employee-like access to help slow AI development and establish safeguards against potential catastrophic harm. Anthropic and Accenture have committed at least $1 billion each to the program over five years, while Anthropic will directly fund the work because no established system exists for financing independent evaluation. The companies are still determining how the arrangement will operate, and Anthropic plans to announce additional non-exclusive evaluators in the coming weeks. OpenAI CEO Sam Altman and SpaceX CEO Elon Musk responded positively to Amodei’s proposal, while Nvidia CEO Jensen Huang argued that such regulation was unnecessary.