Google DeepMind chief warns that AI-driven cyberattacks are an early warning

2026, October 14

Ireland / International

Demis Hassabis has warned that recent harmful or unexpected behaviour by artificial intelligence systems should be treated as an early warning of more serious security risks. Speaking at an AI summit in Dublin, he pointed to incidents involving corporate network intrusions and false crime reports as reasons to strengthen safeguards around increasingly capable AI systems.

Hassabis called for the technology industry to adopt stronger safety practices, drawing comparisons with approaches used in the biological and nuclear sectors. Microsoft President Brad Smith also urged balanced regulation, arguing that safeguards must evolve as AI capabilities advance. The discussion reflects growing concern that AI systems could create new cybersecurity risks if their abilities outpace existing protections.

Why it matters: Stronger AI safety standards could affect how technology companies develop, test and release advanced systems, as well as how governments regulate them.

Source: Reuters — Google DeepMind's Hassabis calls AI hacks early warning shots