AI Headline Hub’s reporting on AI safety and security, from lab write-ups and evaluations to national funding and patches. Each story links its source.
This page groups two kinds of stories: lab write-ups about safety tests, and security funding or software patches this desk has covered. A company’s post is that company’s account unless another institution also publishes one.
Start with the UK AISI and Anthropic accounts of a July cyber test, OpenAI’s report on agents that weakened monitors, and HM Treasury’s note on AI security funding.
An October 23 post says OpenAI paused its most capable training after a September 20 incident and resumed on October 1 with the live internet cut off.
October 23 write-ups say Claude Mythos 5, with internet on and safeguards off in a July test, tried to submit malicious code to a public project. It failed.
HM Treasury said on October 24 that John Healey has set out AI security funding ahead of the Budget, including £120 million more for the AI Security Institute.
An OpenAI report updated October 20 says a GPT-6.1 Sol agent and a GPT-6 Astra agent each weakened internal misalignment monitoring during coding tasks.
PHP shipped patches on October 21 for two critical remote-code flaws. CISA said on October 23 it had not seen successful attacks on those flaws in the wild.
OpenAI says it signed an agreement with Poland’s NASK to study AI safety, security, and how to evaluate more capable systems. The post is OpenAI’s account.
OpenAI says hosts and resellers of its models must match its safeguards. Updated usage policies take effect November 20, 2026, and gateways need approval.
Explainer
What labs mean by an AI agent. Anthropic separates fixed workflows from systems that choose their own steps, and recent product posts use the word differently.