Anthropic Issues Major Threat Report Documenting Widespread AI Misuse for Weapons, Cyberattacks

Anthropic on Tuesday published its most comprehensive threat intelligence report to date, revealing how its Claude AI models were deployed for biological weapons research, sophisticated cyberattacks, and surveillance operations between December 2025 and August 2026. The company identified approximately 35 "distinct research efforts" with concerning activity over a 30-day period, blocking five specific case studies of researchers attempting to use Claude for biological weapons development, including a request for assistance with gain-of-function research on the chikungunya virus.