Dutch security coordinator NCTV says attackers now use AI to carry out cyberattacks, not only to prepare them

National Coordinator for Security and Counterterrorism (Netherlands) · Research report

Read the source (opens in a new tab)
What happened
The Netherlands' National Coordinator for Security and Counterterrorism (NCTV) published its Cyber Security Assessment Netherlands 2026 on November 6, 2026, in Dutch, covering August 2025 to August 2026.
Key points
  • For the first time, the assessment says, attackers demonstrably used AI in carrying out attacks, mostly in a supporting role; it adds that they use AI to write and refine malware.
  • It describes malware that contacts an AI model during an attack to write new commands or rewrite its own code in real time.
  • It says the average time from finding a vulnerability to its exploitation fell from a few months in 2024 to a few weeks in 2025 and a few hours in 2026.
  • It says AI models will likely eventually be able to operate fully on their own, and that attacks would then come at a speed and scale that human action alone likely cannot stop.
Evidence
Annual government threat assessment. It says the gap between the cyber threat and Dutch organizations' resilience is a serious risk to national security. Points are paraphrased from the Dutch text.

See it in the latest record