Artificial intelligence and machine learning techniques have become ubiquitous in recent years, with their adoption across diverse application domains continuing to grow. While AI/ML solutions provide unprecedented capabilities in automation, pattern recognition, and predictive modeling, their widespread deployment raises fundamental questions regarding security, trust, and adversarial robustness. At the same time, the field of computer security is confronted with increasingly sophisticated threats that challenge traditional defense mechanisms, making state-of-the-art AI techniques essential for detecting, analyzing, and mitigating evolving attacks, such as malicious software.

The Security and Artificial Intelligence Lab (SAIL) of the Security and Privacy Unit at TU Wien investigates the intersection of these two rapidly advancing fields by addressing both the opportunities AI offers for security and the risks faced by AI systems themselves. A particular focus of our work lies in analyzing AI models using explainable artificial intelligence (XAI) techniques, as these models often operate as black boxes. By gaining insights into their internal decision-making processes, we aim to identify and address weaknesses in current models and to automatically uncover previously unknown threat patterns.

If you would like to learn more or explore potential collaborations, please reach out to Prof. Daniel Arp.

---

Main directions of our research include:

• Cryptography for Privacy
  Zero-knowledge proofs and blockchains
• Provably secure and efficient cryptography
  Design of cryptographic schemes and optimization for specific applications
• Scalability and Interoperability of Blockchains
  Payment channels and blockchain protocols
• Static analysis
  Frameworks for automatic soundness checks of programs
• Web security
  Formal verification of web protocols and standards
• Secure Systems Lab
  Applied systems security (mobiles, IoT) and malware analysis and detection

• back