Kevin Eykholt

Kevin Eykholt is a Research Staff Member at the IBM T.J. Watson Research Center. He earned his Ph.D in computer science at the University of Michigan Ann Arbor. During his time there, he designed one of the first physical adversarial attacks on computer vision classifiers and object detectors through the use of small adversarial stickers. These stickers, when placed on road signs, caused computer vision systems to mislabel or ignore objects it previously recognized.

Now at IBM, Kevin studies the security of machine learning algorithms and how to use them in security. Adversarial attacks, though widely believed to be a huge threat, appear to be of little concern in real deployments. Kevin believes this is due to impractical defensive measures and unrealistic attack models. Thus, he studies how to mitigate such attacks using low cost methods, while also looking to improve adversarial threat modeling. Kevin is also interested in the security of LLMs given their widespread use and expanded threat landscape. He has been studying the new risks such as bias, jailbreaks, hallucinations, etc. and working on methods to measure, mitigate, or positively exploit these risks in real deployments.