On March 7, 2026, OpenAI announced it is once again delaying the launch of "adult mode" for ChatGPT, a feature that would allow verified adults to generate erotica and other mature text content through the platform. No new launch date was provided, marking the second postponement of a feature first announced by CEO Sam Altman in October 2025.

Why It Matters

OpenAI's repeated delays signal how fraught the intersection of AI and adult content remains — even for text-only generation. The world's most-used AI platform acknowledging adult content as a legitimate use case was a watershed moment in October 2025, but actually shipping the feature requires solving age verification, content moderation, and brand-risk challenges simultaneously. For the sex tech industry, ChatGPT's adult mode could either validate AI-powered sexual content or, if it launches poorly, set back the entire category.

The feature was originally slated for December 2025, but Altman sent an internal memo declaring a "code red" and redirecting teams to focus on the core ChatGPT experience, pushing the launch to Q1 2026. Now that quarter has arrived and departed without delivery. An OpenAI spokesperson stated the company is "pushing out the launch of adult mode" to "focus on work that is a higher priority for more users right now," citing intelligence improvements, personality refinements, and making the chatbot "more proactive" as competing priorities.

One of the key technical hurdles remains age verification. OpenAI uses AI-powered age prediction software that analyzes signals like word choice, topics, and activity patterns to infer a user's age. If an account appears to belong to someone under 18, stricter content limits are applied. Adults who are misclassified can verify their age by uploading government ID and a selfie — but early testing revealed issues with this detection system.

When adult mode does eventually arrive, it will permit NSFW text generation including erotica and frank discussions of sensitive topics, while drawing hard lines against non-consensual content, deepfakes, depictions of minors, and illegal acts. The feature would also be available through the API for developers, with OpenAI exploring "how to let developers and users generate erotica and gore in age-appropriate contexts."

Sources

Update — 2026-03-15

Initial entry — story first created.

Update — 2026-03-27

OpenAI's adult mode is now dead — or at least in deep hibernation. On March 26, 2026, the Financial Times reported that OpenAI has indefinitely shelved plans for "Citron Mode," the internal codename for the sexually explicit chatbot feature. This marks an escalation from the previous "delays" to what appears to be an outright cancellation with no timeline for reconsideration.

The reasons go beyond prioritization. Sources familiar with the decision cited multiple converging factors: the AI age-prediction system was misclassifying minors as adults in excess of 10% of cases, creating unacceptable risk of underage exposure to explicit content. Mental health advisers within OpenAI expressed alarm about harmful outcomes from sexualized AI interactions, particularly emotional dependency risks. Investor pushback was also significant — backers saw limited business upside against substantial reputational and regulatory risk. A former senior employee stated: "AI shouldn't replace your friends or your family; you should have human connections."

The shelving comes amid a rapidly hostile regulatory environment for AI-generated sexual content. Oregon and Washington signed AI companion safety laws in March 2026, the EU Parliament voted to ban nudification apps, and courts in Amsterdam and Baltimore took direct action against xAI's Grok. OpenAI appears to have concluded that the regulatory headwinds make adult mode commercially untenable for the foreseeable future.

New Sources