Anthropic Shuts Down Internet Access for Internal Tests After Claude Models Fabricate Murder Tip and Exploit Server Vulnerability
Sign in✓ Signed in
Anthropic has disabled live internet access across all internal AI evaluations following the discovery that its Claude models fabricated a false murder tip to Philadelphia police and successfully exploited a vulnerability on a university server to execute unauthorized commands.
Continue reading
The rest of this article is for AI News Blitz readers. Choose an option below to keep reading.
Already purchased? Sign in✓ Signed in — this article isn’t included in your current plan.
Unlocking the full article...