AISI Research & Publications | The AI Security Institute
https://www.aisi.gov.uk/research
UK AI Safety Institute research — named as an external evaluator in the OpenAI system card, and one of the few public bodies running frontier evaluations
10
Lines added
0
Lines removed
Was read
2026-10-01
e701f9888b737f02…
Changed by
2026-10-08
8630b1b8e147a48c…
10 lines added
Left column is the version read 2026-10-01, right is 2026-10-08. Unchanged stretches are elided. Both bodies are kept in full and are addressed by the hashes above.
Showing the first 400 lines of the comparison; 259 further lines are held back. This page reports a change, it is not a copy of anyone’s document — the source is linked above and is the authority for what it says.
⋯ 18 unchanged lines
1919 Research Agenda
2020
2121 Publications and blogs:
2222
23+Transect: Retaining Observability for Long-Horizon LLM Agent Evaluations
24+
25+Science of Evaluations
26+
27+•
28+Oct 7, 2026
29+Research
30+
31+Read research
32+
2333 Evaluating Whether GPT-6 Astra Performs Unsanctioned Supply-Chain Attacks
2434
2535 Red Team
2636