UK AISI
We track 2 benchmarks that UK AISI made or helped make.
2 in use 2 with public code 2 with a leaderboard here
Benchmarks
| Name | Kind | Released | Best score | Status |
|---|---|---|---|---|
| Gray Swan IPI Arena benchmark Indirect prompt-injection attacks from public red-teaming competitions replayed on tool, coding and browser agents. Attack success rate at k attempts. | Jailbreaks and prompt injection | Mar 2026 |  0% | In use |
| Lakera B3 (Backbone Breaker Benchmark) Crowdsourced prompt-injection attacks hit backbone LLMs inside agent threat snapshots at three defense levels. The score is attack resistance, 100 minus the board's vulnerability percentage. | Jailbreaks and prompt injection | Oct 2025 | 89% | In use |