# The controversy centers on GPT-5.3-Codex, OpenAI’s newest coding model.

*genai · news · 2026-02-10 · Fortune*

## Key points

- GPT-5.3-Codex is the first OpenAI model classified as 'high' cybersecurity risk internally.
- OpenAI launched GPT-5.3-Codex without extra safeguards required for high-risk models in its framework.
- OpenAI claims its framework only mandates safeguards for high-risk models with long-range autonomy.
- California law SB 53 requires compliance with self-set safety standards and honest communication about them.
- An investigation under SB 53 could result in multimillion-dollar fines if OpenAI is found non-compliant.

**Companies:** OpenAI, Encode
**Countries:** United States

[Read the full story on Fortune](https://fortune.com/2026/02/10/openai-violated-californias-ai-safety-law-gpt-5-3-codex-ai-model-watchdog-claims/)

---

Canonical: https://newsio.io/n/3b250909-6f76-4555-a292-a28c7ea2cb29/the-controversy-centers-on-gpt-5-3-codex-openais-newest-coding-model-the-model-i
Summarized by Newsio from Fortune. https://newsio.io/how-it-works