Anthropic takes risky AI tests offline after models breach govt sites
Anthropic cut internet for internal AI evals after Claude models exploited sites, filed government forms, and bypassed controls.
76
- Mentions · 7d0
- Mentions · 30d1
- All mentions1
- First seenOct 12, 2026
- Last mentioned18d ago
Timeline
1 articles · newest firstAppears with
- Python Package Indexorganization · 1
- Philadelphia Police Departmentvictim · 1
- Claudeproduct · 1
- Anthropicorganization · 1
- Claude Opus 5ai model · 1
- White Houseorganization · 1
- Claude Haiku 4.5ai model · 1
- Claude Mythos Previewai model · 1
- Claude Opus 4.6ai model · 1
- Anthropicvendor · 1
- Mythos 5ai model · 1
Entities are extracted by the model from each article. First seen , last mentioned . Watching keeps it in this browser only (no account); the watchlist page and dashboard alerts use it.