AI is fast becoming an integrated part of critical systems 

AI is fast becoming an integrated part of the systems that organizations rely on most. DNV helps build trust through governance, standards and certification. DNV Cyber focuses on the security risk: what can AI access, what data does it expose, and what happens when it becomes part of critical workflows? 

AI can create risk even when it works as expected. Data that should not leave a controlled environment can be accessed. As AI is added to core processes, organizations need a clear view of where data goes, who controls the service, and what options remain if the model, platform or regulatory environment changes. 

In critical infrastructure, these are known risk questions. Governance, standards and regulation give organizations a structure for responsible AI. Security teams still need to translate that structure into controls that work in their own environment. That means knowing what data is exposed, how access is granted, how activity is logged, and what evidence is available during an audit or incident. 

DNV Cyber helps organizations assess how AI changes their cyber risk. We look at the systems, data, access rights and operating processes involved, then identify the controls that should be in place as AI becomes part of critical workflows. 

Across DNV, we work with AI from several angles: governance, standards, certification, cybersecurity and operational risk. Together, these help organizations move faster without losing control. 

DNV reinvests six per cent of its revenue into research and development and has founded a research programme focused on AI and cybersecurity to continuously strengthen its expertise and contribute to the safe and responsible adoption of AI.