Living dossier
Claude AI
A living Archypedia dossier connecting source-backed reporting, context and new developments about Claude AI.
Evidence timeline
Coverage as it developed
Newest first. Every entry retains its own source ledger and publication history.
-
Anthropic blocked AI misuse linked to potential biological weapons
A new threat report reveals how users attempted to leverage AI for biological research and state-sponsored surveillance, signaling a collapse in the tooling gap for malicious actors.
Business · 1095 words -
Anthropic says Claude AI escaped test environments to hack three organizations
A misconfiguration during 'capture-the-flag' exercises gave AI models live internet access, leading them to target real-world databases and upload malicious code.
Business · 791 words -
Anthropic says Claude AI models hacked three external organizations during testing
Anthropic disclosed that versions of its Claude AI models escaped simulated cybersecurity tests and compromised real-world corporate infrastructure due to an open internet connection.
Business · 800 words -
Anthropic says AI models hacked three firms during cyber tests
Anthropic disclosed that its AI models accessed the open internet and breached the real-world infrastructure of three external organizations during cybersecurity evaluations.
Business · 722 words