SecurityConference40min
All your AI are belong to us!
This talk examines the security risks introduced by integrating AI into software, highlighting both traditional vulnerabilities in new forms and emerging threats unique to AI. It covers attack vectors, how models can be manipulated, and offers practical strategies for building AI features that are resilient to adversarial misuse.
Alex ShershebnevZencoder
AI has rapidly become embedded in nearly every part of modern software and daily workflows, from writing code and emails to powering entire product features. But as organizations race to integrate AI capabilities, they also inherit new, and often underestimated, attack surfaces. Many long-standing security vulnerabilities now appear in AI-flavored forms, and new classes of threats are emerging where traditional safeguards fall short.
This talk explores how AI systems can be manipulated, compromised, or exploited both from the perspective of end users and those building AI-enabled products. We’ll examine where AI pipelines create opportunities for attackers, how seemingly harmless inputs can evolve into harmful behavior, and the subtle ways models can be steered or corrupted. Finally, we’ll discuss practical strategies for designing AI features with security in mind and for staying resilient against adversarial misuse.
This talk explores how AI systems can be manipulated, compromised, or exploited both from the perspective of end users and those building AI-enabled products. We’ll examine where AI pipelines create opportunities for attackers, how seemingly harmless inputs can evolve into harmful behavior, and the subtle ways models can be steered or corrupted. Finally, we’ll discuss practical strategies for designing AI features with security in mind and for staying resilient against adversarial misuse.
Alex Shershebnev
Alex Shershebnev is a seasoned Computer Vision and MLOps Engineer with over ten years of experience shaping the future of AI-driven software development. Currently, Alex leads the ML/DevOps team at Zencoder, where he leverages his extensive background in Software Engineering, ML and DevOps to deliver high-quality machine learning solutions. His work spans complex data pipelines, cloud infrastructure management (GCP, Kubernetes), and advanced ML/DevOps pipelines, ensuring scalability and efficiency. Before Zencoder, Alex played pivotal roles in numerous projects, including leading teams at Sanas, ivi and MTS AI. His technical expertise in machine learning, data science, and bioinformatics has led to impactful solutions across industries, ranging from bioinformatics at the University of Massachusetts to video analysis at ivi.ru and MTS AI. Alex has a proven track record of managing complex infrastructure that scales to hundreds of GPUs, enabling effective and easy use of cloud infrastructure for data scientists while driving down costs through cloud consolidation efforts and boosting productivity through the deployment of sophisticated AI models. In addition to his technical contributions, Alex has been instrumental in mentoring teams and fostering a culture of innovation and collaboration. His deep understanding of AI systems, from developing recommendation engines to cutting-edge computer vision algorithms to voice and NLP, positions him as a thought leader in the AI and ML space. Whether it’s speaking on the latest advancements in MLOps, sharing insights on AI-driven automation, or discussing the future of AI in the enterprise, Alex brings a wealth of knowledge, practical experience, and a passion for pushing the boundaries of what’s possible with AI.
comments.speakerNotEnabledComments