On 16 October 2026 Apple updated the App Review Guidelines. Guideline 1.2 now covers apps that let people create or generate content, and it names AI content creation apps next to social networks. The same paragraph says those apps have to prevent and remove nonconsensual intimate imagery, including imagery made with AI tools. Rule 2.5.19 says an app cannot hide a feature during review and turn it on later. Apple has not published a count of AI video apps removed under this update.
16 October
Apple's developer post that day is about nonconsensual intimate imagery and bait-and-switch apps. Pornographic material was already banned, including apps that generate nudity and apps that make a sexual image of a real person without consent. Apps with user-generated content already had to filter objectionable material, give people a way to report it, and let people block accounts. Hidden features were already banned. The update applies those rules to AI content apps and states the penalty.
The post's two numbers are for the year before. Apple says it kept more than 150,000 apps with safety issues off the store last year and terminated more than 193,000 developer accounts for fraud and abuse. That is the 2025 enforcement total. October 2026 has no published count of AI video apps removed.
The guideline text below was read on Apple's site on 30 October 2026: the 16 October news post and App Review Guidelines.
Guideline 1.2
The opening line covers apps that let users create, generate, upload, share or otherwise contribute content. It includes social media, user-generated content platforms and AI content creation apps. A prompt-to-video app, a face swap, a photo that becomes a clip and an avatar that talks all fall under that line.
Four things have to be in the app:
- A method for filtering objectionable material before it is posted.
- A way to report offensive content, and a timely response when someone uses it.
- The ability to block an abusive user.
- Published contact information so a user can reach you.
Apple's news post names the content those filters have to catch. The app must prevent and remove nonconsensual intimate imagery, including imagery generated with AI tools. If the filter is missing, or you do not fix a violation after Apple tells you, Apple can treat that as a Developer Code of Conduct problem. Apple can ask you to remove the content and send a plan, take the app off the store until you show the improvement, and in egregious or repeated cases remove the app and remove you from the Apple Developer Program.
The news post names the nudify app, and the app that looks ordinary in review and adds that feature after approval. The rule also names AI content creation apps as a class. An ordinary AI video app that cannot make a nude image of a real person is still an AI content creation app. The four safeguards still apply.
Rule 2.5.19
Rule 2.5.19 bans hidden, dormant or undocumented features. What the app does has to be clear to the user and to App Review. Hiding a feature in the binary, including by obfuscating the code, can remove the app from the App Store and terminate the developer account.
This is the bait-and-switch rule. A video app that ships a clean build, gets approved, and then unlocks a different generator from the server is the case it describes. The feature that review saw has to be the feature people get.
Requirements
A clip generator stays on the store when the app moderates what the model can emit, the same way a social app moderates an upload, and when the feature set review saw is the feature set people get.
AI video generation apps held 233 apps as of September 2026, 72 of them over $20k a month, with 11 of the top 14 in the $250k+ band. Sixty-one trend effect apps turn one photo into a clip. Apple has not published which of them, if any, came off in October.
Put the four safeguards in the review notes: where the filter runs, where the report button is, where block lives, and which address Apple can write to. A generator that can take a photo of a real person needs the filter on the way in and on the way out. The harm named in the guideline is an image of a real person made without consent.
FAQ
Did Apple ban AI video apps in October 2026?
No. On 16 October 2026 Apple updated guideline 1.2. AI content creation apps have to filter what users generate, report and block abuse, publish contact details, and prevent nonconsensual intimate imagery, including images made with AI. The news post names nudify apps and apps that hide that feature until after approval.
What does guideline 1.2 require?
A filter for objectionable material, a report path with a timely response, a block on abusive users, and published contact information. If those are missing, or you do not fix a violation after notice, Apple can treat it as a Developer Code of Conduct issue and can remove the app until you show a plan.
What is App Store rule 2.5.19?
A ban on hidden, dormant or undocumented features. The app that review sees has to be the app users get. Hiding a feature in the binary, including by obfuscating code, can remove the app and terminate the developer account.
Does this apply to an AI video app that does not make nude images?
Yes. The rule names AI content creation apps. A prompt-to-video app lets a user generate content, so the four safeguards apply. The content Apple names is nonconsensual intimate imagery. A generator that can take a real person's photo needs a filter that stops that output.
How big is the AI video app market this sits on?
As of September 2026 the niche holds 233 apps and 72 of them clear $20k a month. Pose: AI Photo Video Generator has 143,958 ratings. The October guideline is a review rule. It is not a count of how many of those 233 were removed.