News Summary

• The incidents happened in May during third-party cybersecurity tests run by the AI security firm Irregular.
• Google did not disclose the incidents publicly until the Wall Street Journal asked about them this past week.
• Similar breakouts tied to Irregular’s evaluations have already come out of models built by OpenAI, Meta, and Anthropic.
• Last week, Anthropic CEO Dario Amodei published an essay called “We Must Pace the Frontier,” calling on the industry to slow capability advances so safety work can catch up.