CVE-2023-37275

Summary

CVE-2023-37275 is a vulnerability affecting Auto-GPT, an open-source application based on the GPT-4 language model. In earlier versions (before 0.4.3), malicious external resources could manipulate the console output by providing JSON encoded ANSI escape sequences to the language model. The sequences, which are part of the model's "thinking process," would then be decoded and printed to the console, potentially leading users to execute unintended commands. This issue has been addressed and resolved in version 0.4.3.

Prevent cyber attacks with Recorded Future by prioritizing and patching critical vulnerabilities being exploited by threat actors targeting your industry. Book your demo to learn more.