Detecting and Countering AI Misuse: Report

A threat intelligence report reviews misuse cases against Claude-series models between Dec 2025 and Aug 2026 across seven harm areas—including cyber operations, influence ops, scams, and biological misuse—and describes disruptions and safeguards applied.

• Anthropic: adversaries used AI to target US naval forces2026-09-16
• Anthropic merges Claude chat and Cowork UI2026-09-16
• Google opens smart home to third‑party AI agents2026-09-16
• HN discussion: release age and training cutoff for 20 models2026-09-16
• A warning about 'model welfare'2026-09-16