What changed on 16 October
Apple published a changelog and a separate post explaining the reasons. Three changes matter for an app that generates images:
- Guideline 1.2 now says it covers “social media, user-generated content platforms, and AI content creation apps”, and names nonconsensual intimate imagery among the problems those apps must deal with. Apple’s post says apps must proactively prevent and remove such imagery, including imagery made with AI.
- Guideline 2.3.14 (new) sets the rules for the new creative assets. Product page headers and search result images must relate to your app, suit all audiences, and leave out unverifiable claims and specific or discounted prices.
- Guideline 2.5.19 (new) forbids hidden, dormant or undocumented features. Apple says it clarifies a long-standing policy, aimed at apps that behave one way in review and another after approval.
The four things guideline 1.2 asks for
Guideline 1.2 lists what these apps must include. Apple’s wording is on the left; on the right is how we would apply it in an image app.
| Apple asks for | In an AI image app, that could mean |
|---|---|
| A method for filtering objectionable material | Checks on the photo that goes in and on the image that comes out, not only on a text prompt. |
| A way to report offensive content, with timely responses | A report action on every result, and someone who reads the reports. |
| The ability to block abusive users | A way to suspend an account that keeps trying to misuse the service. |
| Published contact information | A support address in the app and on the product page. |
Apple’s post adds that developers who do not put adequate safeguards in place, or do not fix violations after notice, risk breaking the Developer Code of Conduct and being removed from the App Store.
Why templates make this easier
An app built around templates rather than a free text prompt has an easier job here. The range of possible results is narrower, so filtering is simpler and surprises are rarer. It is one of the reasons none of our photo apps has a prompt box, which we wrote about in another post.
Consent still comes first
The update sits on top of a rule from November 2025: guideline 5.1.2(i) requires a clear disclosure and explicit permission before personal data, such as a photo, is shared with a third-party AI service. In JobShot, that is a photo privacy dialog before the first upload. Tapping Cancel means no photo leaves the phone.
A checklist for this week
- Filtering on what goes into the model and on what comes out of it
- A report action on results, and a process for answering reports
- A way to suspend accounts that abuse the service
- Contact details in the app and on the product page
- Disclosure and permission before a photo goes to a third-party AI service
- Nothing in the build that is switched off for review
- Header and search result assets without prices or claims you cannot back up
Every change in this update, and the earlier ones this year, is in our guide to App Review Guidelines changes.
Sources: Apple’s App Review Guidelines, and its posts of 16 October 2026, Updated App Review Guidelines now available and Strengthening App Review Guidelines against nonconsensual intimate imagery and bait-and-switch apps.
Related: App Review Guidelines changes · AI app development · Photo app development