Enhancing Software Security with GPT-5.2-Codex: The Next Evolution of Enterprise AI Coding

In a recent update, OpenAI introduced GPT 5.2, along with enhancements to its coding model Codex, expanding its capabilities for agentic use cases. GPT-5.2-Codex is described as the most advanced agentic coding model for real-world software engineering, optimized for long-horizon work with agents and stronger cybersecurity features.

This model, a branch of GPT-5.2, focuses on agentic building, offering support for complex tasks in software engineering and cybersecurity. OpenAI stated that GPT-5.2-Codex advances the tools available for responsible security research, aiding developers and defenders in tackling intricate, long-term projects.

Enterprises can now access the new Codex model through various platforms for paid ChatGPT users, with plans to extend access to API users soon. OpenAI is also piloting a program with trusted users to explore more permissive models for cybersecurity work, balancing accessibility and safety.

Enhancements in Cybersecurity

OpenAI highlights GPT-5.2-Codex as its most robust cybersecurity model to date, emphasizing the need for a deployment strategy that accommodates future growth and supports defensive cybersecurity.

The company tested the model on benchmarks such as Capture-the-Flag (CTF) evals, CVE-Bench, and Cyber Range, where GPT-5.2-Codex excelled in various scenarios, showcasing its capabilities in cybersecurity tasks.

Cybersecurity Deployment Project

Users of GPT-5.1-Codex-Max, introduced in November, identified a source code exposure vulnerability in React using the model, prompting OpenAI to focus on improving cybersecurity features in subsequent models. The company aims to strike a balance between deploying cutting-edge models and ensuring robust tools for defensive cybersecurity.

Advancements in Agentic Frontiers

GPT-5.2 has been praised for its utility in business tasks, and with the Codex version, these capabilities are expected to expand, particularly in coding for agents. The model’s improvements in long-horizon work and performance on extensive code changes make it a valuable tool for enterprises.

Since its preview release in May, Codex has played a significant role in promoting agentic and vibe coding in the enterprise AI builder space, offering enhanced capabilities in generating and managing coding projects. The platform has transitioned from simple code completion to supporting complex coding tasks seamlessly.