Apple has tightened App Store moderation so developers must actively prevent and remove AI-generated nonconsensual intimate imagery, with apps containing such content facing removal from the storefront. The new rules also explicitly target apps that hide their true purpose during review and switch on nudification features after approval.
What the rule change covers
The updated App Review guideline 1.2 names AI content creation apps and obliges developers to filter out deepfakes. Sexually explicit content was already prohibited, but the new language spells out that generative AI outputs fall under the same ban and that developers cannot wait for complaints before acting. Apple warns that offending apps can be pulled from the App Store, and that Apple Developer Program membership can be terminated.
Guideline 2.5.19 is also new. It clarifies that any app that adds features or capabilities not present in the version submitted for review risks removal. Apple links the rule directly to nudification apps that present as harmless image editors during the review process but provide nonconsensual imagery tools only after approval, a pattern the company labels bait-and-switch.
Why the rule was added now
Pressure from consumer complaints about AI chatbots generating nude and sexualized deepfakes, including reports involving xAI’s Grok app, pushed Apple to act. Earlier in 2026, Apple informed xAI that Grok would be removed from the App Store if the app continued to allow nude deepfakes. xAI introduced safeguards, and Apple approved an updated version of the app. The new guidelines formalize what had been an ad hoc enforcement decision into a written rule that applies to every developer.
How Apple handles hidden features
Apple’s developer notice calls out apps that misrepresent their purpose during review and only reveal harmful capabilities after approval. Under guideline 2.5.19, an app that ships a working nudify tool it never disclosed to reviewers can be removed, and the developer behind it can be terminated from the Apple Developer Program. Apple’s stated position is that it does not hesitate to remove apps and terminate developers that violate these guidelines.
Changes to alternative app distribution
Apple also updated section 5 of the guidelines to extend its ban on apps that facilitate human trafficking or child exploitation to alternative app distribution channels. The change closes a gap that previously allowed prohibited content to circulate through third-party app marketplaces that operate under Apple’s alternative distribution framework on devices in the European Union and elsewhere.
What developers need to do
- Audit AI generation pipelines for any pathway that can produce nonconsensual intimate imagery of real people and remove it.
- Make sure the build submitted for App Review matches what ships in production, with no dormant nudify or deepfake features turned on after approval.
- Review section 5 obligations for any app distributed through alternative app marketplaces, not only the public App Store.
FAQ
What did Apple change in the App Store rules about AI deepfakes?
Apple expanded guideline 1.2 to explicitly cover AI-generated nonconsensual intimate imagery and added guideline 2.5.19, which targets apps that hide capabilities such as nudification during App Review and enable them after approval. Apple also extended section 5’s ban on human trafficking and child exploitation content to alternative app distribution channels.
Can apps be removed from the App Store for AI deepfakes?
Yes. Apple states that apps with offending content can be removed from the App Store and that developers who violate the guidelines can be terminated from the Apple Developer Program. The company has already taken this step with at least one AI app that permitted nude deepfakes, requiring safeguards before an updated version was approved.
What are nudify or nudification apps?
Nudification apps are image editing tools that generate sexually explicit depictions of real people without their consent. The new guideline labels the tactic of concealing this functionality during App Review and switching it on after approval as bait-and-switch, which is grounds for removal under guideline 2.5.19.