Proposers
Tiago H. Falk and Marina L. Gavrilova
Workshop Code
Please use the following code when submitting your paper to this Workshop: W12-SECURITY
Workshop Web Page
Please, follow this link to see the latest C4AI4C information.
Scope and Aims
The 2026 Workshop on Cybersecurity for AI & AI for Cybersecurity (C4AI4C), formerly known as the Workshop on AI for Cybersecurity: Friend or Foe?, is in its 3rd edition following two successful workshops in Honolulu, Hawaii (2023) and Vienna, Austria (2025). The Workshop aims to bring together experts from academia, government labs, and industry to discuss the advantages and challenges of using AI systems for cybersecurity and security in general. On one side of the equation, machine and specifically deep learning have been shown to achieve state-of-the-art results across numerous applications, particularly those related to anomaly/intrusion detection, biometrics, and computer vision, to name a few. On the other side, known vulnerabilities, such as data poisoning, adversarial attacks, hallucinations, social engineering, and deep fakes, make their widespread adoption dangerous, especially for safety-critical applications and those dealing with privacy-sensitive information. These two sides need to be considered when developing models of trust for AI-based systems and agents.
Content and Objectives
The use of AI in cybersecurity and security is a topic of extreme importance worldwide. It is, however, a double- edged sword that needs to have researchers working on both sides of the spectrum working together. This workshop will enable that. In this workshop, we will bring researchers working on both sides of the problem, i.e., on the use of AI for cybersecurity and the cybersecurity issues of AI, to the same room to brainstorm ideas on how to move the field forward in the most efficient way possible. The workshop will be comprised of posters, full-paper technical presentations, invited speakers, and networking events. These topics are of extreme relevance not only to cybersecurity and security researchers, but to the global AI community.
Invited Speakers and Workshop Structure
This will be a FULL-Day Workshop. The Workshop will combine four invited talks (two from academia, one from industry, one from government) and a mixture of posters and full papers. Full papers will be submitted to CAI 2026 and will undergo peer review. Posters will be submitted as abstracts to the organizers and will not appear on IEEE Xplore. Projector and screen will be needed for full papers and poster boards for posters. For the invited speakers, these are not yet fully defined but speakers that are leaders in the field will be invited. The audience should be knowledgeable about AI algorithms (e.g., transformers, generative AI) and some of their vulnerabilities (e.g., adversarial attacks, deep fakes, backdoor attacks). Presenters who have papers accepted will be experts in these fields, as their papers will be peer reviewed.
Previous Workshops
Previous editions of the “Workshop on AI and (cyber)security: Friend or Foe?” were organized by Tiago Falk and colleagues at the IEEE SMC 2023 and 2025. The website and programs of these first two editions can be found at: https://sites.google.com/view/smc2023workshopcyberai/ and https://sites.google.com/view/smc2025workshopcyberai/.
Prof. Gavrilova started the International Conference Series ICCSA in 2003, the conference is now approaching its 26th edition The 26th International Conference on Computational Science and Its Applications. She also co-organized other workshops, most recently The 4th Explainable AI for Computer Vision (XAI4CV) Workshop at CVPR 2025.