Repository navigation
|
How can a team catch Generative AI output without moderation or safeguards automatically before a build goes to review? |
Replies: 1 comment
|
The quick check. Run this from the project root and read what it prints. grep -rni 'api.openai.com\|anthropic\|generativelanguage\|stable diffusion\|text-to-image' . && ! grep -rni 'moderation\|report\|block user\|content filter' .Read the result honestly. This is a heuristic. A match is something to inspect, not proof of a violation, and an empty result means nothing was found, not that the app is compliant. For CI, run the full guard, which reports this as What the hit means. A generative AI or image generation integration is present without content filtering, reporting, blocking, age rating, or abuse safeguards. Apple applies UGC rules to AI output. Google has a dedicated AI Generated Content policy. Since 16 October 2026 Apple 1.2 names AI content creation apps explicitly and requires them to prevent and remove nonconsensual intimate imagery, including AI-generated imagery (https://developer.apple.com/news/?id=vvzx6cul). The fix. Add moderation, reporting, blocking, an accurate age rating, and abuse safeguards. Prevent NSFW, deepfake, face swap, and undress generation. Recipe source. data/detection-recipes.json This is a maintainer FAQ for app-store-compliance, an open playbook and pre-submission guard for App Store and Google Play review. Run |
The quick check. Run this from the project root and read what it prints.
Read the result honestly. This is a heuristic. A match is something to inspect, not proof of a violation, and an empty result means nothing was found, not that the app is compliant. For CI, run the full guard, which reports this as
BOTH-AI-GENERATED-CONTENTwith a coverage line and a proper exit code.What the hit means. A generative AI or image generation integration is present without content filtering, reporting, blocking, age rating, or abuse safeguards. Ap…