Archive — US — techcrunch
UK drops ‘safety’ from its AI body, now called AI Security Institute, inks MOUwith Anthropic
The UK has rebranded its AI Safety Institute as the AI Security Institute, shifting its focus from existential risks to enhancing cybersecurity in response to AI threats. A partnership with Anthropic aims to explore AI's role in public services, aligning with the government's strategy to boost the economy through technology and modernization despite concerns over AI safety and oversight.
Key points
- The UK government renamed the AI Safety Institute to AI Security Institute, shifting its focus to cybersecurity.
- The initiative aims to address national security risks posed by AI technologies.
- A new MOU with Anthropic will explore incorporating its AI assistant Claude into public services for efficiency.
- The government emphasizes modernization and economic growth through AI, sidestepping previous existential safety concerns.
- Civil servants will have their own AI assistant called 'Humphrey,' promoting use of AI for efficiency.
- Despite changes, officials claim the institute's commitment to addressing serious risks remains intact.
Why it matters
- This shift aligns with global trends emphasizing AI's role in enhancing national security and governance.
- Historically, the UK's AI policies have focused heavily on safety and ethical considerations; this change marks a significant pivot.
- As of 2025, there is ongoing global discussion on balancing AI development with safety regulations, particularly evident in the U.S.
- The move mirrors a broader trend among governments to harness AI for economic growth while grappling with associated risks.
The renaming of the AI Safety Institute to the AI Security Institute signifies a strategic pivot in UK policy, prioritizing economic enhancement and cybersecurity over previous safety concerns, a shift reflective of the evolving global landscape of AI governance.