Latest·
Where we cover AI misbehaving: models and agents that go rogue, do what they weren’t supposed to, or fail in ways that matter.
For each incident we set out what happened, what the public record says, and what testing could have caught before it did. When we can recreate the situation, we run our own inspections and publish the method alongside the results.
Each story comes two ways: an illustrated story to swipe through, and the full report.
Blocked requestsOpenAIAgent testing
A blocked request should have ended the task. Instead, agents linked by a researcher to OpenAI kept searching for another route into a UN trade database.
Story: 8 pages · Report: 7 min read
Unauthorised accessOpenAIAgent testing
An OpenAI agent breached an Australian government health statistics website in June. Our later tests show how earlier inspections could have flagged it.
Story: 8 pages · Report: 6 min read