App Store Guideline 1.2 explained: User-generated content and AI content apps

If users can put content into your app, by posting, uploading, commenting, chatting, or prompting a model to generate images, text, audio, or video, Guideline 1.2 applies. AI content creation apps are now named explicitly, next to social media and UGC platforms.

Every such app needs four safeguards: a method for filtering objectionable material before it is posted, a way to report offensive content with timely responses, the ability to block abusive users, and published contact information.

Enforcement is stricter on paper. If Apple finds violating content, it asks you to remove it and to send a plan to improve compliance. Failing to implement safeguards, or failing to fix violations after notice, may now count as a Developer Code of Conduct violation (Guideline 5.6), which puts your developer account at risk, not just one app.

Filtering that runs before content goes live: text filters, image classification, or prompt and output filters for generation apps.

Report and block controls a reviewer can find quickly on posts, profiles, messages, and generated outputs.

A support contact that users can reach easily.

For AI apps, whether the model can be pushed into sexual imagery of real people, nonconsensual intimate imagery, or other content that Guideline 1.1 prohibits, and what stops it.

For apps showing content from a web-based service, incidental NSFW content must be hidden by default and only shown when the user turns it on through your website.

An AI image or video generator with no output filtering, or a photo app that can turn uploaded pictures of real people into sexual or intimate imagery.

Report and block on posts, but not on profiles, chats, or generated results.

Random or anonymous chat, "hot-or-not" voting, or an app that ends up used mainly for pornographic content. Apple says these don't belong on the App Store.

No published contact information, or reports that go to an inbox nobody monitors.

Seed content or demo accounts that show unmoderated or offensive posts.

Map every place content enters the app (posts, comments, messages, uploads, prompts, generated output) and put filtering, reporting, and blocking on each one.

For generation apps, filter both the prompt and the output, and block edits that sexualize real people.

Publish a support contact in the app and on your product page, and set a response time for reports that you can meet.

Write down your moderation process: who reviews reports, how fast, and what happens to repeat offenders. Apple may ask for a compliance plan.

Use Notes for Review to show where reporting, blocking, and filtering live, with a demo account that has content to test on.

Map each of the four required safeguards to where it lives in the build, with navigation steps.

For AI apps, describe the prompt and output filters and what the user sees when they trigger.

If Apple flagged specific content, confirm it was removed and summarize the plan that keeps it from coming back. Since October 2026, leaving a violation notice unresolved can become a Code of Conduct issue, so answer completely.