Artificial intelligence could lower the barriers keeping biological weapons out of terrorists’ hands, Britain’s MI5 chief warned Tuesday.

Sir Ken McCallum’s warning came with an important qualification: that nightmare scenario has not been realized. His concerns center on what increasingly capable AI might enable in the future.

This places a critical question before governments and the technology industry: How can powerful AI’s benefits be preserved while preventing its misuse for mass murder?

In MI5’s annual threat update, McCallum noted terrorists have already used AI to enhance surveillance and attack planning. He also highlighted risks posed by AI-enabled drones. The biological concern specifically addresses potential future capabilities—not confirmed attacks—emphasizing that no such weapon currently exists in terrorist hands as stated by the intelligence chief.

McCallum announced the launch of an AI Threat Assessment Centre within MI5, unifying expertise from intelligence, AI safety, law enforcement, and cybersecurity to distinguish credible threats from exaggerated claims.

The warning underscores urgent preparation needs. It must avoid inflating future capabilities into current realities. Recent data shows MI5’s work against state-sponsored threats has quadrupled over the past decade, now matching the scale of its counterterrorism efforts. The agency is also prioritizing national resilience protections for infrastructure, society, and critical systems like electricity, water, and financial networks.

Cybersecurity specialists caution that gaining access to a company network does not automatically grant control over essential machinery—yet such access creates exploitable vulnerabilities. Protecting these systems requires concrete actions: transparent testing, clear vulnerability disclosures, and accountability for repairs—not just assurances of safeguards.

McCallum’s alert demands practical responses focused on measurable outcomes, not theoretical risks. Governments must ensure critical services withstand disruption while balancing technological progress with public freedoms. The standard for action must be tangible protection, verifiable results, and unwavering respect for individual rights.