MI5 launches AI Threat Assessment Centre to separate hype from national-security risk

On 13 October 2026, MI5 Director General Sir Ken McCallum used his annual threat update at Thames House to announce that the Security Service has launched the AI Threat Assessment Centre (AITAC) during 2026.

What AITAC does

Housed in MI5, the centre draws expertise from the UK AI Security Institute (AISI), the National Crime Agency and GCHQ’s National Cyber Security Centre. McCallum said it produces all-source assessments that “separate the bluster from the realities” so ministers can seize AI opportunities while managing misuse risks at pace. MI5 positions AITAC alongside earlier joint analysis structures for terrorism and state threats.

Risks McCallum highlighted

In the same speech, McCallum said AI is already intensifying the cybersecurity contest, can tailor propaganda and radicalising content to individuals, and has been sought by terrorists to improve reconnaissance and attack methods, including AI-enabled drones. He attached greatest weight to AI’s not-yet-realised potential to change the stakes on existing or new bioweapons. He also flagged the longer-term alignment problem — facing an actor that could out-think humans — as genuinely novel and deserving serious attention inside MI5.

Context

The announcement sits inside a broader elevation of MI5’s Protect mission for national resilience. Official AITAC pages confirm the centre’s role: multi-agency, all-source assessments of AI national-security threats for government departments. Independent reporting (POLITICO, BBC) summarised the same launch and bioweapon warning.