Domestic intelligence chief warns AI could become a threat in its own right

Simone Smit, head of the Dutch intelligence service AIVD, warns that artificial intelligence could eventually pose a threat in its own right, beyond its use to facilitate existing threats. She made the remarks on Wereldupdate, a podcast by Dutch think tank Clingendael.

Smit says that "AI is also being used as a geopolitical instrument. It offers incredible opportunities, but it is also a genuine threat that we are investigating. We are increasingly seeing that it may even constitute a threat in its own right."

The AIVD and other security services have previously warned that artificial intelligence could help malicious actors carry out large-scale cyberattacks. "AI is making attacks bigger and more complex, and they will become more frequent," the agencies said in their earlier warning.

The agencies said AI can help attackers identify and exploit vulnerabilities before organisations even realise their systems are at risk. The technology also lowers the technical barrier to carrying out attacks, allowing a wider range of malicious actors to launch cyber operations. The agencies urged organisations to prioritise cybersecurity, review whether their defences remain adequate and take AI-related warning signs seriously.

Dutch intelligence agencies have previously warned that AI is becoming a factor in geopolitical competition, particularly as countries seek technological advantages with potential military applications.

In a December 2024 assessment, the AIVD, the Military Intelligence and Security Service (MIVD) and the National Coordinator for Counterterrorism and Security (NCTV) warned that shifts in countries' AI capabilities could affect the military balance and create new vulnerabilities for the Netherlands and its allies.

They specifically discussed China and Russia in the context of strategic competition and national security. China has sought to become a global leader in AI, with access to foreign knowledge and technology forming part of that effort.

The 2024 assessment also identified potential uses of generative AI to produce malware, phishing messages, disinformation and propaganda. The agencies said terrorist groups, including al-Qaeda and Islamic State (ISIS), were likely already using generative AI to produce propaganda, potentially increasing its volume and reach.

Simone Smit became director-general of the General Intelligence and Security Service (AIVD) on March 1, 2026. Before that, she served as deputy director-general of the service from 2021. She previously held senior roles at the National Coordinator for Counterterrorism and Security (NCTV), including responsibility for counterterrorism, and worked in police leadership roles in Rotterdam and The Hague.