Adversarial AI in the Cyber Domain

In our From the Author’s Mouth series, we open our virtual doors to the movers and shakers in information security. It allows you to get your information directly from the source in an interactive interview- setting, with sufficient time reserved for Q&A.

This time we welcome Puck de Haan, Scientist Cybersecurity & AI at TNO, and one of the authors of the research paper Adversarial AI in the cyber domain.

Can't make it on this date? No worries, just register and you will automatically receive a link to the video directly after the event....

Adversarial AI in the Cyber Domain

Artificial Intelligence (AI) systems use large amounts of data to make decisions in a complex system. In order for an AI system to learn specialized tasks, such as discrimination of the different elements within the complex system it operates in (also known as classification), Machine Learning is applied. These are computer programmes that learn automatically and efficiently through experience. Proper analysis of the ability of these AI systems to withstand external threats is essential before they can be deployed on a large scale.

What threats are associated with the use of AI? This is a question that TNO seeks to answer through its recent research into the vulnerabilities of AI applications in the cyber domain.

The Adversarial AI field of research studies the vulnerabilities in AI systems. Adversarial AI is a subject of research that has acquired a lot of traction and scientific attention over the past years, with significant results. For example, researchers have succeeded in misleading AI systems that classify images enabling them to have a turtle classified as a gun.

But despite these significant developments, most Adversarial AI research remains focused on the computer vision and text domains. That is why TNO’s research has focused on Adversarial AI applications in the cyber security domain.

Topics:

#What is Adversarial AI?

#What are Adversarial AI attacks?

#How can attackers target AI systems?

#What defensive measures can be taken?

#How can Adversarial AI be applied in cyberspace?

#What conclusions can we draw?

Agenda

14h00: Welcome -- Wim Stoffelen

14h10: Adversarial AI in the cyber domain -- Puck de Haan from TNO

14h35: Q&A Session

14h50: Closing remarks and next webinars -- Wim Stoffelen