OWASP June meetup: Hidden Risks in Open-Source Code and AI Models

Hidden Risks in Open-Source Code and AI Models
Ryan Wakeham, Senior Solutions Engineer, Checkmarx

This presentation will explore some of the threats and risks associated with open-source code and generative AI models. Open-source packages, while widely adopted for their flexibility and collaborative benefits, can be exploited through supply chain attacks, where malicious code is introduced into trusted components. Similarly, open-source AI models face risks such as data poisoning, model theft, and adversarial attacks, which can compromise their integrity and reliability. Understanding these threats is crucial for developing robust security measures and ensuring the safe use of open-source technologies.