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

Source: techcrunch.com · Section: business-tech · Bias: Center · Published: Sat, 10 Oct 2026 00:18:32 GMT

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

Read the full article at techcrunch.com →

The same story across the political spectrum (12 sources)

Center: An Anthropic AI model sent a false homicide tip to Philadelphia police — techcrunch.com
Center: Anthropic bans users from being 'cruel' to its AI systems — www.bbc.com
Center: Anthropic now offers a free vulnerability-finding service for open-source software — www.engadget.com
Center: Anthropic's Claude AI fabricates eyewitness account, submits false murder tip to police website — www.foxbusiness.com
Right: Scoop: AI companies plot "day after" scenarios for public revolt — www.axios.com
Right: Anthropic’s AI Sent Police A False Report Of A Homicide — www.dailycaller.com
Right: DUKE: I Reserve The Right To Call My Chatbot A ‘Retard’ — www.dailycaller.com

← Back to FlashTopics.com