Apple tightens App Store rules on AI deepfakes

Apple has expanded its App Store Review Guidelines to explicitly cover AI-generated nonconsensual intimate imagery, while also tightening the rules around hidden features and apps that misrepresent themselves during review. The changes give Apple clearer grounds to remove apps and developers that fail to act on harmful content. ¹

The headline change is to guideline 1.2. Apple now says content-moderation requirements explicitly include nonconsensual intimate imagery generated with AI. Apps hosting user-generated content must prevent and remove that material, while developers that fail to implement adequate safeguards or respond to violations risk breaching Apple’s Developer Code of Conduct and removal from the App Store.

That wording matters because the rule moves beyond a general ban on pornography. Apple is tying the obligation to proactive moderation: developers are expected to stop abusive material from spreading and to deal with it when it appears. The company’s announcement specifically cites “nudification” apps that create explicit images of real people without consent.

Apple has also added guideline 2.5.19 to clarify its longstanding ban on bait-and-switch behaviour. An app must not present one set of features during App Review and then add hidden or undocumented capabilities after approval. Apple says apps that misrepresent their purpose, including by concealing harmful features until approval, can be removed from the App Store and the Apple Developer Program.

The update extends Apple’s language around Section 5, too. Apps facilitating human trafficking or the exploitation of children were already prohibited, but Apple says the revised wording makes clear that apps containing this material will be removed and blocked from alternative distribution as well.

MacRumors reports that the revised rules could affect apps such as Grok, which has previously faced scrutiny over AI-generated nude deepfakes. That context is separate from Apple’s formal announcement: the company’s developer post does not name Grok, but it does say the updated rules are intended to keep pace with emerging risks and novel user experiences. ²

Apple says it prevented more than 150,000 apps with safety issues from reaching customers last year and terminated more than 193,000 developer accounts for fraud and abuse. Those figures are company-reported, but they show the scale of the enforcement system behind the rule changes.

Our opinion

Apple’s update is less a brand-new ban than a clearer enforcement map. The useful change is that AI-generated abuse is now named directly, removing room for developers to pretend older moderation language did not cover it. The bait-and-switch clarification is just as important: an app cannot pass review as one thing and quietly become something more harmful later. The real test will be consistency. If Apple applies these rules quickly and transparently, they could make the App Store less hospitable to abusive tools without leaving developers guessing about the line. If enforcement remains reactive, the clearer wording will amount to little more than a warning label.