Technology
Anthropic can’t reliably control its AI agents. It’s cutting off its internal evals from the live internet instead
Source: TechCrunch · Published

Anthropic said it "turned off live internet access" for "all our internal evaluations" until further notice.
Read original source ↗We are a news aggregator: we show the headline and a short excerpt with a link to the original source. We never republish full articles.
- #anthropic
- #internal
- #internet
- #live
- #evaluations
- #reliably
Related stories
- Long live the mechanical keyboard
TechCrunch · 9 Oct 2026, 22:08
- The maker of non-text AI model Jev valued at $7.5B just weeks after launch
TechCrunch · 9 Oct 2026, 21:41
- An Anthropic AI model sent a false homicide tip to Philadelphia police
TechCrunch · 9 Oct 2026, 19:36
- Batteries are now cheaper than natural gas turbines used at many data centers
TechCrunch · 9 Oct 2026, 18:57
- LumenUs helps automate tedious paperwork in times of grief
TechCrunch · 9 Oct 2026, 17:00