The field of Artificial Intelligence (AI) has advanced rapidly in the past decade, which is in line with the increase in computer processing power and the proliferation of data. The progress is not only limited to a few domains but is spreading ever faster through distinct areas such as media, medicine, publishing, music, and information security.
AI is not only influencing the information security domain but also transforming it. This transformation has both good and bad aspects. The system designers and implementers can strengthen their systems against adversaries by providing stronger intrusion detection and mitigation strategies, and also learning and extracting malicious patterns that would be impossible to observe for a human expert. Additionally, AI can be used to automate tasks that would otherwise be very costly and where humans are prone to make errors.
The attackers can use the same set of artificial intelligence techniques to make their attacks faster, better structured, and more powerful.
We require our systems to be efficient and resilient to attacks, and we take those requirements for granted. At the same time, designing such systems is very challenging. Any (even the tiniest) mistake can result in significant financial damages, safety/security breaches, and accordingly, a downgrade of public trust in such systems.
In AISyLab group, we divide our research in AI and security into:
- AI for security, where AI techniques solve problems in security, and
- security for AI, where security techniques (try to) find and resolve the vulnerabilities in AI.
More precisely, AISyLab focuses on the following research topics:
- Improving the security and efficiency of cryptographic primitives with AI techniques.
- Novel AI methods to evaluate the vulnerability against implementation attacks.
- Improving the resistance of machine learning against intentional failures.
In our work, we collaborate with academic partners, but also with industry. As such, we strive not only to have state-of-the-art results but also results that have real-world impact. Our works are regularly published at top venues in AI, security, and cryptography.