Anthropic blocked AI misuse tied to biological weapons

In its third misuse report, Anthropic says it blocked attempts to use its models for cyberattacks, surveillance, and biological-weapon‑relevant research, publishing example code and prompts and urging governments and peers to act. It reports finding misuse between Dec 2025 and Aug 2026 and is planning an IPO this fall.

• Anthropic: adversaries used AI to target US naval forces2026-09-16
• Anthropic merges Claude chat and Cowork UI2026-09-16
• A warning about 'model welfare'2026-09-16
• Are Chinese firms handing US data to US firms?2026-09-15
• Anthropic confirms Claude used in weapons and surveillance2026-09-15