OpenAI updated the GPT-5.5 System Card detected on July 30, 2026. The change replaced three featured research and safety articles in the document's header with two different research publications. The prior version linked to articles on 'Safety and alignment in an era of long-horizon models' (Jul 20), 'Why teens deserve access to safe AI' (Jul 16), and 'GPT-Red: Unlocking Self-Improvement for Robustness' (Jul 15). The updated version now links to 'How enabling two settings tripled our scores on the ARC-AGI-3 benchmark' (Jul 29) and 'Scientific computing in the age of agentic AI' (Jul 28), while retaining the safety alignment article. This is a navigation and featured content update with no material change to the system card's substantive governance or safety disclosures.
This change updates the featured research articles linked in the GPT-5.5 System Card but does not modify the substantive safety, deployment, or operational disclosures that govern the model's use. The system card's core governance framework remains unchanged; only the curated list of research references has been updated to highlight more recent publications on AI benchmarking and scientific computing.
Updated from three safety and robustness articles to two AI benchmarking and scientific computing articles while preserving the safety alignment article.
This change record describes what was added, removed, or modified in the document. Analysis reflects what the updated agreement states or permits. It does not constitute a legal determination about enforceability. Applicability may vary by jurisdiction. Methodology
This is a content and navigation update to OpenAI's system card documentation. The change replaces featured article links without modifying the substantive safety disclosures, model capabilities, deployment safeguards, or risk assessments described in the card …
Regulatory exposure, obligation change, escalation trigger, board-ready language, and recommended action for legal and compliance teams.
Unlock the full institutional analysis — InsightConductAtlas provides verified policy intelligence sourced directly from platform documents. All analysis is intended to support, not replace, legal and compliance review. Record CA-C-004043.
OpenAI updated the related-articles section of its GPT-5 System Card detected on July 30, 2026. The document previously linked to …
OpenAI's usage policies document was detected with a language change on July 29, 2026. The language selector and document header …
No material change detected. The sentence describing data usage for model training remains identical in both versions. The policy continues …
H.R. 8094 would make the FTC the referee for AI model disclosure. It also names system cards as a way to comply, which turns a voluntary in…
The bill does not regulate most AI startups directly. But it changes the companies they depend on. Here is what the first federal AI law wo…
Ad personalization controls removed. Contact scanning added. Advertiser data partnerships quietly dropped. A timeline of every change.
Get the research letter
Companies change their terms quietly. We read every version and catch what actually changed. One email a week on the changes that matter and what they mean.
Get alerted when this policy changes again, including what changed and why it matters.