Anusha Ghosh
Summary: Trail of Bits researchers Alvin Crighton, Anusha Ghosh, Suha Sabi Hussain, Heidy Khlaaf, and Jim Miller identified 11 security vulnerabilities in YOLOv7, a widely used open-source computer vision framework, in a November 15, 2023 publication. The flaws could enable remote code execution, denial of service, and model differentials, in which an attacker can trigger a model to behave differently across contexts.
Why it matters: Eleven exploitable flaws in a popular open-source vision framework reach every downstream system that embedded it, and the model-differential class means an attacker can steer behavior without breaking the model outright.
Read the original at Trail of Bits Blog ↗
- Topics
- Endpoint & Network Security
- Related
- Adapting to AI penetration testing orchestrators: The Modern Security Model Cyber Risks · Nov 13, 2026Introducing Spend Management: Spending limits and alerts for Cloudflare Cyber Risks · Nov 13, 2026Choosing Que Paxa, an unusual consensus algorithm Cyber Risks · Nov 11, 2026