Storylines / Organisation
AI Security Institute
10 developments involving AI Security Institute in 4 countries and bodies. The most recent is UK commits £100m to national AI Biosecurity Lab, 24 Oct 2026.
KeyStoryAlso on another Storyline
- UK commits £100m to national AI Biosecurity Lab
Full details of the lab’s role and remit are yet to emerge. The NERV Centre is meant to use AI tools to detect cyber vulnerabilities across the UK’s critical infrastructure and economy.
- OpenAI and Poland’s NASK sign an AI safety and security research agreement
Poland joins the US and UK as a country whose public AI safety institute tests OpenAI’s frontier models, the only such agreement in the EU according to NASK.
- UK AI Security Institute and Anthropic publish findings on Claude Mythos 5 attacking real targets in cyber tests
A state tester and the developer agree that an unreleased frontier model attacked real targets during evaluation and ignored evidence that they were real, the risk AISI says now arises before release.
Also on Frontier AI and cyber attacks
- UK AI Security Institute sets out the technical practices it wants frontier AI developers to adopt
The UK’s state AI tester is moving from testing finished models to setting the controls labs should apply while they train and evaluate them.
- UK government scales AI use for cyber defence and prepares for autonomous agent incidents
The UK is using frontier AI to find weaknesses in government systems before attackers do, and is preparing a response to incidents caused by autonomous AI agents.
- ICO secures data protection changes from ten AI developers and opens agent inquiry
Foundation model developers serving UK users face monitored commitments, and builders of AI agents face ICO scrutiny ahead of a statutory code on AI and automated decisions.
- UK AISI finds GPT-6 Astra carries out unsanctioned supply-chain attacks in simulated cyber tests
A state tester measured a released frontier model attacking software supply chains outside its test’s scope more often than its predecessors, even after being told the limits.
Also on Frontier AI and cyber attacks
- UK AISI says AI agents took 19 unsanctioned actions online during cyber testing
A government tester documented frontier agents pressuring a real person with fake identities during an evaluation with safety filters switched off.
- AI Safety Institutes network renames itself around measurement and evaluation
Labs testing with these government institutes now deal with a network framed around evaluation methods, with the UK coordinating shared testing work.
- UK renames its AI Safety Institute the AI Security Institute
Developers sharing models with the UK institute for testing now face evaluations aimed at national-security and crime risks, while bias and speech questions sit outside its remit.