The partnership focuses on creating embedded evaluation units that will work directly alongside Anthropic’s developers. To execute this, Accenture plans to leverage the expertise of Faculty, an applied AI firm acquired earlier this March. Faculty specializes in the complex testing frameworks required to ensure that large-scale models operate within safe parameters.

This initiative follows urgent calls from top industry figures for standardized safety oversight. Anthropic CEO Dario Amodei recently advocated for a collaborative industry approach to AI evaluation, a sentiment echoed by OpenAI leadership. As Microsoft concurrently rolls out its own updated safety guidelines, the move highlights a shift toward prioritizing defensive infrastructure in the race to deploy increasingly powerful AI systems.