image

Yisroel Mirsky is a tenured Assistant Professor and Zuckerman Faculty Scholar in the Department of Software and Information Systems Engineering at Ben-Gurion University. He is the head of the Offensive AI Research Lab at CBG. He received his Ph.D. from BGU in 2018 and was a postdoctoral fellow for two years in the at the Georgia Institute of Technology in the research labs of Prof. Wenke Lee. His main research interests include AI safety, AI agent security, deepfakes and adversarial machine learning. Dr. Mirsky has published his work in some of the best security venues: USENIX, CCS, NDSS, Euro S&P, Black Hat, DEF CON, RSA, CSF, AISec, etc. His research has also been featured in many well-known media outlets: Popular Science, Scientific American, Wired, The Wall Street Journal, Forbes, and BBC. Some of his works, include the exposure of vulnerabilities in the US 911 emergency services and research into the threat of deepfakes in medical scans, both featured in The Washington Post.

Prospective Students

I am actively seeking excellent M.Sc., Ph.D. and post-doc students who would like to research offensive AI (adversarial machine learning, AI safety, deepfakes, attack automation, …). If you are interested, come check out our lab or send me an email!

Consulting, Research, & Training

With over a decade of experience in red teaming and AI security research and as a principal investigator in over $4 million of projects with government and industry, I provide expert guidance to organizations navigating the complex landscape of modern AI threats. My services are designed for companies seeking to ensure the safety of their AI deployments or to defend against AI-enhanced adversaries.

Services Offered

  • Consulting: Strategic advisory and technical oversight tailored to your mission. I provide the expert review, actionable insights, and robust solutions necessary to navigate complex security and AI landscapes
  • Training: Professional talks and training seminars delivered at all levels. From in-depth lectures covering all aspects of offensive AI to engaging invited talks at corporate events, I provide custom educational experiences tailored to your organizational needs.
  • Research: Independent, cutting-edge investigations into emerging threat vectors and technologies. I offer specialized analysis and deep-domain exploration to stay ahead of the rapidly evolving Offensive AI frontier.

Interested in working together? Please reach out via email at yisroel@pegasec.io to discuss your project or event.

Grants

Over the years, I have been awarded a number of grants from a wide variety of institutions. Working with these organizations has been an incredible experience. Thank you for your support!

Talks

Other Talks:

  • ‘AI Security Forum` (Tel Aviv 2025)
    Offensive AI is here. Now what?
  • Sandia National Laboratories (USA 2025) Keynote Talk
    ML/DL Workshop - Offensive AI: The Dark Side of Intelligent Systems
  • DEF CON 32 (USA 2024)
    Your AI Assistant has a Big Mouth: A New Side Channel Attack
  • BIRD - US & Israel Dept Energy - ICRDE (Webinar 2023)
    The Threat Horizon of Real Time Deepfakes
  • CSIRO Data61 (Australia 2022)
    The Threat Horizon of Deepfakes
  • RSA Security Conference (USA 2021)
    Securing Tesla \& Mobileye From Split-Second Phantom Attacks
  • Royal Holloway University of LondonInvited Talk – Research Seminar (2019)
    Medical Deepfakes: How malware can automatically tamper CT and MRI Scans
  • MIT - Massachusetts Institute of Technology Research Seminar (2019)
    An invited talk on my research at the ALPHA group of CSAIL (2019).
  • DEF CON 27 AI Village (USA 2019)
    Automated Injection \& Removal of Medical Evidence in CT and MRI Scans
  • Black Hat Asia (Singapore 2019)
    Briefing: See Like a Bat: Using Echo-Analysis to Detect Man-in-the-Middle Attacks in LANs
  • Keio University 9th International Cybersecurity Symposium (Tokyo 2019)
    The Security of AI
  • Bell Labs Research Center, Nokia (2018)
    Online Anomaly Detection Algorithms for Securing the Internet of Things
  • Keio University 7th International Cybersecurity Symposium (Tokyo 2018)
    Panelist: The Future for the Security of AI
  • DBMI Data Mining and Business Intelligence (Israel, 2017)
    Securing IoT Video Surveillance Systems with Online Machine Learning
  • DeepSec In-depth Security Conference (Austria, 2016)
    Bridging the Air-Gap – Data Exfiltration from Air-Gap Networks
  • CODE BLUE (BlackHat Japan) Security Conference (Tokyo, 2016)
    Air-Gap Security: State-of-the-art Attacks, Analysis, and Mitigation

Publications

Below is a list of selected publications. For an up-to-date list of publications, please see my Google Scholar.