MI5 Director General Sir Ken McCallum has launched a new AI Threat Assessment Centre to combat risks including the potential for terrorists to develop biological weapons. Speaking at Thames House, he warned that artificial intelligence is already supercharging the cybersecurity contest
between attackers and defenders.
The AI Threat Assessment Centre Launch
The new centre will assemble a team of teams
from a range of bodies to produce assessments of risks. Sir Ken McCallum described the initiative as a team of teams
approach, designed to address the unique challenges posed by systems that are increasingly capable of out-thinking human actors. He emphasized that humanity has no historical experience in facing an actor capable of out-thinking us, a situation he deemed genuinely novel.
The announcement follows a September 22 commitment by the Prime Minister to ensure the UK remains at the forefront of secure AI development. The AI Security Institute (AISI) is also working alongside industry partners to establish technical standards, noting that risks are now emerging earlier in the development lifecycle—during AI development itself—rather than just at the point of public release. The AISI coordinates the Network for Advanced AI Measurement, Evaluation and Science, contributing to international taxonomy standards.
Terrorist Misuse of Biological and Drone Technology
During his annual threat update, Sir Ken identified the potential for AI to assist in the creation of bioweapons as the misuse case to which I’d attach greatest weight.
He defined these as disease-causing agents, including bacteria, viruses, fungi, or toxins, used to deliberately harm people, animals, or crops. While he noted that such capabilities are not yet realised, he warned that terrorists are already seeking to exploit AI to improve reconnaissance and attack methodologies, specifically citing the use of AI-enabled drones.
The threat is compounded by the internet’s role in radicalization. He noted that extreme right-wing terrorism continues to skew young, driven by the internet.
.png "Photo: The AI Security Institute (AISI)")
Ministerial Measures and Cyber Defenses
These include strengthening cyber protections and preparing for incidents involving autonomous AI systems.
Minister Kanishka Narayan stated that the government intends to legislate wherever necessary to ensure Britain remains one of the hardest countries to attack using AI. The government is also using the Frontier AI Lab, established in partnership with Thomson-Reuters, which focuses on the safety, reliability, and societal impacts of new technology. The Minister argued for a whole-society approach to resilience, ensuring that technology continues to serve British interests and prosperity.
NEW: MI5 chief warns terrorists could use AI to develop biological weaponsMI5 director general Sir Ken McCallum has warned that artificial intelligence could allow terrorist organisations and hostile states to develop dangerous biological weapons, including those capable of… pic.twitter.com/wjRFA931X0
Resilience in Post-9/11 Security
Addressing how intelligence agencies prepare for the novel threat of super-intelligent actors, Sir Ken drew a parallel to the period following the 9/11 attacks. He recalled that after the 2001 terror attacks, the agency faced thousands of well-trained jihadis without having a confident answer on how to tackle the threat. He noted that MI5 will take inspiration from that era of tackling al Qaeda as it faces opaque risks from artificial intelligence.
Sir Ken concluded by emphasizing that the British public has a vital role to play in maintaining national resilience as threats become more complex.
