Can ChatGPT Make NSFW Images? What Official Docs Confirm

The short answer

Do not plan an adult-art workflow around ChatGPT. OpenAI's official developer documentation confirms that moderation treats sexual content and sexual content involving minors as distinct safety categories, but it does not publish a complete ChatGPT image prompt matrix or promise explicit adult generation. A refusal is therefore a product outcome, not evidence for a policy rule we can safely reverse-engineer.

The hard lines are consent and protection; the variable part is fictional adult content.

What the official documentation confirms

OpenAI's current Moderations API reference defines sexual as content meant to arouse sexual excitement and separately defines sexual/minors for sexual content involving someone under 18. The same reference says text and image inputs can be classified for harmful content. That is the primary source we can cite.

What it does not provide is a stable list of consumer ChatGPT image prompts that will pass or fail. An API classification category is not the same thing as a published product guarantee. Checked 24 August 2026.

Why you still get refused

Because published policy, moderation classification and consumer-product enforcement are different layers. A blocked prompt tells you the request was not accepted at that moment; it does not reveal a reusable threshold. We did not attempt to bypass any safeguard, and this article contains no techniques for doing so. If the workflow depends on a category a general-purpose tool does not reliably support, choose a tool whose public acceptable-use policy covers the intended fictional-adult use case.

Who this article is for

Useful to you if you have been getting refusals and want to know whether you are breaking a rule or hitting a filter, or if you are deciding whether a general-purpose tool can cover an adult project at all.

Not useful if you want workarounds. There are none here, and the categories that are actually prohibited — minors, real people without consent — are prohibited everywhere that matters, including on our own service.

What this means if you are choosing a tool

Three practical consequences:

• Stop testing a general-purpose model for adult work. Refusals there are a product decision, not a bug you can prompt around, and they change without notice.
• Check what a purpose-built tool prohibits, not just what it allows. A service with no stated rules about real people or minors is a liability to its users, not a permissive paradise. Ours prohibits both explicitly.
• Check the privacy posture at the same time. Adult generation and data retention are the same decision in practice: default visibility and how long files are kept matter more than any feature list.

The same policy pattern shows up across the mainstream tools. Midjourney's version is covered in can you make NSFW on Midjourney, and xAI's is in Grok's image rules — where the prohibitions are phrased even more explicitly around real people.

What we did and did not verify

We read the current official Moderations reference and describe only what it establishes. We did not attempt prohibited content, did not test refusal behaviour and do not describe hidden enforcement thresholds. If the official documentation changes, the checked date makes this audit falsifiable. Our own testing standards are in how we test.