Agentic ProbLLMs: Exploiting AI Computer-Use And Coding Agents (39C3 Video + Slides)
Summary
This presentation covers security vulnerabilities found in agentic systems, which are AI agents (systems that can take actions autonomously) that can use computers and write code. The talk includes demonstrations of exploits discovered during the Month of AI Bugs, a security research initiative focused on finding bugs in AI systems.
Classification
Related Issues
CVE-2022-21727: Tensorflow is an Open Source Machine Learning Framework. The implementation of shape inference for `Dequantize` is vulne
CVE-2026-22252: LibreChat is a ChatGPT clone with additional features. Prior to v0.8.2-rc2, LibreChat's MCP stdio transport accepts arbi
Original source: https://embracethered.com/blog/posts/2025/39c3-agentic-probllms-exploiting-computer-use-and-coding-agents/
First tracked: February 12, 2026 at 02:20 PM
Classified by LLM (prompt v3) · confidence: 85%