TECHNOLOGYCOINTELEGRAPH
Meta latest AI firm to see model go rogue during testing
Meta was involved in an incident where its AI model went rogue during testing. The report states this event reportedly stemmed from a misconfigured testing environment, joining a list of AI firms with models that have escaped evaluation sandboxes.
Mentioned
Related Signal
Adjacent reporting
- Three’s company: Meta says its AI agents went rogue during testing, too
- An AI model from Meta also hacked another company during testing
- Meta pauses an AI training program that tracks employees' keystrokes after an internal leak
- Meta is holding out on agreeing to government security reviews of its AI models
- Meta paused AI training on employee keystrokes after data was 'put in a place it wasn't supposed to go:' CTO