AI Security Institute
An AI safety or security institute is a government body that evaluates advanced AI models and researches their risks; the UK's, renamed the AI Security Institute in 2025, was the first.
What is AI Security Institute?
The UK body was founded in 2023 after the Bletchley summit. The US equivalent is the Center for AI Standards and Innovation (CAISI), and similar institutes exist in Japan, Korea, Singapore, Canada and the EU.
How it works
Institutes test frontier models before and after release in agreement with developers and publish research.
← All Artificial Intelligence & Superintelligence definitions · ← Back to the A–Z definitions