Simon Willison found that OpenAI's Codex API—a code-writing tool the company built years ago—still grants access to GPT-5.5, a more advanced model than the public GPT-4o release, despite no official announcement or changelog documenting this capability.
The Codex endpoint appears to be a legacy interface that was never fully deprecated; users calling it receive GPT-5.5's responses without needing to pay for a separate model tier or join a waitlist, bypassing OpenAI's usual product gatekeeping.
For developers and businesses relying on OpenAI's APIs, this means some teams may already have access to a newer, more capable model than they realize—but also raises questions about whether OpenAI intended this exposure or will close it, potentially breaking existing integrations that depend on the Codex endpoint.
Ask the AI about this article →
For example, today's edition would include:
AI-summarized, only the topics you pick — one digest a day via Email, Slack, or Discord.
Free · takes 30 seconds · unsubscribe anytimeWhat is AIToday? →
Ask AI anything about this article. Q&As are published on this page for other readers too.
Sandisk says its NAND-based High Bandwidth Flash (HBF) technology can match HBM bandwidth while providing eigh…
World Labs unveiled Atlas, an omni-model trained on text, images, video, and 3D data that anchors every input…
Saudi Arabia's LEAP tech conference opened with $15 billion in planned technology investments
Nvidia invested $3.5 billion in MediaTek, a Taiwanese chipmaker, to help customers build custom AI chips that…
Semafor and Riddance AI uncovered a network of about a dozen YouTube channels using real actors with AI-genera…
At a recent ICRA panel, robotics researchers discussed how to handle the overwhelming number of publications
