Anthropic and Accenture commit $2 billion to independent AI model testing

Anthropic and Accenture announced Friday a partnership to invest at least $2 billion combined over five years in independent evaluation of Anthropic's frontier AI models, amid rising safety concerns. The investment aims to build capacity for this work. The partnership comes as regulators and researchers pressure AI developers to ensure advanced models are safe. Anthropic CEO Dario Amodei called on companies to slow frontier model development and grant independent evaluators greater access, while rival OpenAI began publishing reports on concerning model behavior. Accenture's AI unit Faculty will lead evaluations, including red-teaming and alignment assessments Anthropic's "embedded evaluation" model places independent evaluators inside companies with employee-level access, allowing them to verify safety commitments and identify blind spots. Evaluators can report incidents and inform the public on risks and benefits, with plans to work with other developers similarly