GPTracker: A Large-Scale Measurement of Misused GPTs
Xinyue Shen, Yun Shen, Michael Backes, Yang Zhang
IEEE Symposium on Security and Privacy 2025 · Day 1 · LLM Security
In an era where large language models (LLMs) are rapidly evolving into sophisticated agents capable of browsing the web, executing code, and integrating with external tools, the emergence of LLM-specific application marketplaces presents both innovation and significant security challenges. This talk, "GPTracker: A Large-Scale Measurement of Misused GPTs," presented by Xinyue Shen, delves into the dark side of this new ecosystem, specifically focusing on the OpenAI GPT Store. The research introduces **GPTracker**, a novel framework designed for the continuous, large-scale collection and analysis of custom GPTs, aiming to identify and understand the prevalence and operational mechanisms of those designed for malicious purposes.