Anthropic tabs Accenture as embedded evaluator to help with AI slowdown proposal
The partnership is non-exclusive, and Anthropic expects to announce other evaluators in forthcoming weeks with which it will also work.
Anthropic has selected Accenture as its first embedded evaluator to assist in slowing down the pace of AI development, as proposed by the company's CEO Dario Amodei last week. Amodei published a three-step plan on September 12 to regulate AI growth and implement safeguards, following warnings of potential catastrophic consequences from rapid, uncontrolled development.
OpenAI CEO Sam Altman and SpaceX CEO Elon Musk expressed their support for the proposal, although Nvidia CEO Jensen Huang disagreed, arguing that such regulation is unnecessary. Amodei emphasized that AI has been advancing at an unprecedented speed, driven primarily by AI's ability to build the next generation of AI systems, which could quickly surpass human understanding and control if left unchecked.
Anthropic has committed to independent evaluators with employee-like access, a move it announced on Friday by partnering with Accenture and its AI business Faculty. These evaluators will help assess and red-team models, conduct alignment assessments, and test model safeguards. The specifics of this partnership are still being developed, as embedded evaluation is a relatively new concept.
Both Anthropic and Accenture plan to invest at least $1 billion in the project over the next five years. The partnership is non-exclusive, and Anthropic anticipates announcing additional evaluators in the coming weeks with whom it will collaborate. However, there is currently no established system for funding independent evaluation, so long-term financial support may come from pooled sources or government funding. In the meantime, Anthropic will directly fund Accenture's efforts.
Written by urgent.news from Cointelegraph's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
