OpenAI patents AI agents that scan code changes for security flaws
The filing describes a system that uses large language models to automatically check new code updates for security weaknesses. It works by creating a specific 'threat model' for the software and using an AI agent to compare the new code against that model to identify potential risks.
In the filing’s own words
The present disclosure relates to systems, non-transitory computer-readable media, and methods for analyzing vulnerabilities within computer code. The disclosed systems can retrieve, by a generative artificial intelligence (AI) system, a commit from a codebase storing commits defining computer code changes saved for a computer application. In one or more cases, the disclosed systems can generate, utilizing a large language model, a threat model comprising a structured data object defining vulnerability parameters of computer code corresponding to the computer application. Additionally, the disclosed system can generate, by analyzing the commit and the threat model together utilizing a generative threat analysis agent, a threat object indicating threat data associated with the commit.
What exactly they claim (claim 1)
1. A computer-implemented method comprising: retrieving, by a generative artificial intelligence (AI) system, a commit from a codebase storing commits defining computer code changes saved for a computer application; generating, utilizing a large language model of the generative AI system, a threat model comprising a structured data object defining vulnerability parameters of computer code corresponding to the computer application; and generating, by analyzing the commit and the threat model together utilizing a generative threat analysis agent, a threat object indicating threat data associated with the commit.
Facts
- Company
- OpenAI (OpenAI OpCo, LLC)
- Official title
- DETECTING VULNERABILITIES IN CODE COMMITS UTILIZING GENERATIVE MODEL AGENTS
- Publication number
- US20260350179A1
- Application number
- 19364425
- Filed
- Oct 21, 2025
- Published
- Nov 5, 2026
- Inventors
- 3
- Claims
- 20
Classified under
G06F 21/577 Physics & computingG06F 8/71 Physics & computingH04L 41/16 Electricity & electronicsG06F 2221/033 Physics & computingMore from OpenAI in Security & privacy
- OpenAI patents AI agents that find security flaws in code Nov 5, 2026
- OpenAI patents AI system to find security flaws in computer code Nov 5, 2026
- OpenAI patents system that hides passwords from AI agents Nov 3, 2026
Summaries are simplified from the filing’s own abstract and first claim. A patent filing is not a product announcement.
Patent data and drawings: U.S. Patent and Trademark Office (USPTO). Nebula is not affiliated with or endorsed by the USPTO.
