Question to the Cabinet Office:

To ask the Minister for the Cabinet Office, what assessment his Department has made of the biosecurity risks arising from increasingly capable artificial intelligence models; and what safeguards are in place to assess and mitigate those risks before advanced models are deployed.

The AI Security Institute works proactively with frontier laboratories to test the capability of AI models and advise them on where safeguards can, and should, be strengthened to prevent biological risks emerging.

The UK Biological Security Strategy sets out our ambition and approach to building national resilience to a spectrum of biological risks by 2030. This includes strengthening our sovereign capabilities to respond to naturally occurring pandemics, biological weapons and emerging biosecurity threats, such as those exacerbated or accelerated by AI.

The UK Biological Security Strategy Implementation Report, published in July this year, sets out the tangible actions we are taking to reduce biological risks, including around £1bn of investment in health protection measures to strengthen domestic resilience.

Earlier this year, the Ministry of Defence announced an investment of £115 million to raise the UK’s defences against the threats from AI, including using AI to improve our biological security and mitigate risks posed by autonomous AI agents. This is alongside investment of over £1.83 billion to establish a Network of National Biosecurity Centres, including £1 billion in Weybridge, £580 million in Defence Science and Technology Laboratory (Dstl) Porton Down, and £250 million over this Parliament in Harlow to bolster defences against biological incidents, accidents and attacks.