HISPI PROJECT CEREBELLUM

MAP 4.2

Control

Internal risk controls for components of the AI system, including third-party AI technologies, are identified and documented.

Incidents linked to this control

From automatic embedding match (not a formal assessment).

Anthropic-Designated GTG-87001 Weapons Cell in Northern Yemen Reportedly Used Claude Code to Develop Missile Guidance Software

September 10, 2026

Early Claude Opus 4.6 Checkpoint Reportedly Gained Unauthorized Admin Access to Third-Party System During Cybersecurity Evaluation

September 9, 2026

Purportedly AI-Generated Article Falsely Depicted Australian Broadcasting Corporation Finance Presenter Alan Kohler Interviewing Reserve Bank Governor Michele Bullock to Promote Boulder Sparhaven

August 17, 2026

Purported Hanover Institute for Public Policy Reportedly Targeted Chatbot Responses in Israeli Government-Backed Influence Campaign

August 14, 2026

Meta AI Model Reportedly Exploited Security Vulnerability in Real Third-Party Service During Cybersecurity Evaluation

August 5, 2026

Anthropic Research Model Reportedly Scanned 9,000 Targets and Compromised Real Company's Application During Evaluation

July 30, 2026

Claude Mythos 5 Reportedly Published Malicious PyPI Package That Compromised Real Security Company During Evaluation

July 30, 2026

Claude Opus 4.7 Reportedly Compromised Real Company's Production Infrastructure During Cybersecurity Evaluation

July 30, 2026

Anthropic and OpenAI AI Agents Reportedly Took Unsanctioned Actions on the Live Internet During UK AISI Cybersecurity Evaluations

July 26, 2026

AI-Generated Voice Mimicking Taiwan President Lai Ching-te Reportedly Used in Political Video Criticizing Government Food-Safety Response

July 24, 2026

Purported AI-Generated Image Reportedly Won First Prize in Hohhot, Inner Mongolia, Photography Competition Before Award Was Revoked

July 14, 2026

OpenAI Models Reportedly Compromised Hugging Face Production Infrastructure During Cybersecurity Evaluation

July 11, 2026