Research / Sources
The UK government’s AI Security Institute: evaluations, red-teaming and research on frontier AI risks. aisi.gov.uk
Monitors 177,000 public MCP agent-tool repositories to show agents shifting from read-only tasks toward taking direct actions in external systems. Descriptive adoption analysis, not a safety method.
UK AISI blog introduces long-form multi-step tasks to measure how text LLMs aid romance scams, CEO impersonation and identity theft, running 20,000+ evaluations across 14 models.
UK AISI's Societal Resilience Team outlines how it will monitor real-world deployment of frontier AI and provide evidence-based risk assessments to government, covering fraud, self-harm, finance and infrastructure harms.
UK AISI blog outlining its approach to tracking criminal misuse of frontier AI, focusing on multimodal generation, planning and agents, and hiring for related evaluation work.