The first institution of its kind in the world. The United Kingdom will allocate millions.

The UK government is establishing a new AI safety laboratory, described as the first of its kind globally, aimed at addressing emerging threats. Minister Kanishka Narayan emphasized the importance of combining scientific knowledge to protect citizens as AI evolves. The lab will monitor risks ranging from cyberattacks to biological threats. OpenAI and Anthropic are set to support the lab with £200 million in credits for their models, though terms are not yet disclosed. Additionally, £120 million will be allocated to the UK Institute for AI Safety, with further funds for projects enhancing cybersecurity. Recent vulnerabilities in PHP have raised concerns, prompting the government to temporarily disable certain services as a precaution.