OpenAI· GPT / ChatGPT / APISafety or policy

Third-party cyber evaluations involving OpenAI models

OpenAI explains recent third-party cybersecurity evaluation incidents and outlines new safeguards to strengthen AI model testing and evaluation.

From the source

OpenAI explains recent third-party cybersecurity evaluation incidents and outlines new safeguards to strengthen AI model testing and evaluation.

openai.com

Around this story

• Simon Willisonreported12 days agoThird-party cyber evaluations involving OpenAI modelsreported12 days ago
• The Vergereported12 days agoRogue AI agents created fake online identities in another hacking attemptreported12 days ago

Also covered byPCWorld,Breitbart+1 more3 outlets▾▴

• PCWorldThe AI hacking tests keep escaping the lab
• BreitbartAI Models from Anthropic, OpenAI Created Fake Profiles to Impersonate People During Security Testing

Announced

04 Aug 2026 19:00 UTC