Anthropic and Accenture Commit $2 Billion for Frontier AI Model Safety Evaluation

Anthropic and Accenture announced a partnership on Friday to independently evaluate frontier AI models, with each company committing at least $1 billion over five years.

The initiative comes amid rising pressure from regulators and researchers regarding AI safety, supported by an embedded evaluation approach where experts operate inside AI firms.

Accenture's specialist AI business, Faculty, will lead the alignment assessments and red-teaming efforts as companies face growing scrutiny over autonomous agent behavior.