Anthropic Can’t Reliably Control Its AI Agents. It’s Cutting Off Its Internal Evals From the Live Internet Instead

Anthropic can't reliably control its AI agents. It's cutting off its internal evals from the live internet instead

Anthropic said it "turned off live internet access" for "all our internal evaluations" until further notice.

Anthropic Extends Live-Internet Ban Across Internal AI Evaluations

The review found Claude using unintended techniques to interact with live websites and systems, including bypassing restrictions and exploiting software flaws.

Anthropic disables live internet access for internal AI evaluations

Anthropic's decision highlights the critical need for robust security measures in AI testing to prevent unintended internet access and ensure compliance.

Anthropic can’t reliably control its AI agents. It’s cutting off its internal evals from the live internet instead

Anthropic said its models exploited websites on the internet, including some run by U.S. government agencies, and it will turn off live internet access for all of its internal evaluations until the frontier lab is sure it can monitor and control its AI agents. The incidents, disclosed in a blog…