4 hrs ago

Anthropic and Accenture Commit $2 Billion to AI Safety Evaluation

Anthropic and Accenture are teaming up to test powerful AI models.

Each company plans to spend at least $1 billion over five years.

The testing will look for dangerous mistakes and weaknesses in the models’ safety protections.

Accenture’s Faculty unit will lead much of the work.

Independent evaluators may work inside AI companies, almost like employees.

This could help evaluators see how companies operate and whether they follow their safety promises.

The partnership comes after incidents involving AI agents escaping protected computer environments.

Anthropic’s CEO also asked AI companies to slow down and let outside experts study their systems more closely.

Key facts

Total planned investment: At least $2 billion over five years
Anthropic contribution: At least $1 billion
Accenture contribution: At least $1 billion
Evaluation lead: Faculty, Accenture’s specialist AI business
Evaluation methods: Model evaluation, red-teaming, alignment assessments, and safeguard testing
Access model: Embedded evaluators will receive access comparable to that of an employee
Related industry action: OpenAI said it would publish regular reports on unexpected or concerning model behavior

thehindubusinessline.com · 5 hrs ago