San Francisco Orders Apple, Google to Pull Nudify Apps

For months, apps designed to strip clothing from photos of real people using generative AI sat quietly in the same app stores as banking apps and productivity tools. On July 17, 2026, San Francisco City Attorney David Chiu tried to change that, sending formal demand letters to Apple and Google that order both companies to remove 13 "nudify" apps within 28 days or face civil penalties under California law.

Key Takeaways

• San Francisco City Attorney David Chiu sent demand letters to Apple and Google on July 17, 2026, targeting 13 AI nudify apps: 8 on the Apple App Store and 5 on Google Play.
• The letters cite California Civil Code Section 1708.86, which lets victims of nonconsensual deepfake pornography sue not just creators but companies that knowingly facilitate or recklessly aid and abet its creation.
• Apple and Google each have 28 days to respond or the city attorney's office says it will pursue civil penalties against them directly.
• Apple says it has removed three of the flagged apps and is terminating their developer accounts; Google says it has suspended all five apps named in the letter.
• This is not Chiu's first move against deepfake platforms: his office sued 16 websites offering similar nonconsensual image generation tools in 2025.

What Are Nudify Apps?

Nudify apps use generative AI to digitally remove clothing from photos of real, identifiable people without their consent, producing synthetic nude images the subject never posed for. Chiu's investigators found the 13 apps his office targeted were not marketed under that label directly. Instead, most listed themselves in the app stores as generic "face swap" or photo editing tools, a framing that appears designed to slip past content moderation while the underlying feature, generating nonconsensual sexual imagery, remained available once installed.

The office's investigation identified 8 such apps on the Apple App Store and 5 on Google Play, all violating both companies' publicly stated policies against nonconsensual sexual content, according to the letters. Chiu did not name the apps publicly, and this article does not either, in keeping with standard practice for reporting on tools built to enable image based abuse.

What Legal Basis Does San Francisco Have?

San Francisco is relying on a California law that treats knowing facilitation of nonconsensual deepfake pornography as its own violation, not just the act of creating the images. Civil Code Section 1708.86, originally passed as AB 602 in 2019 and expanded effective January 1, 2026, gives victims a civil cause of action against anyone who creates and knowingly shares sexually explicit deepfakes of a real person without consent, and against those who knowingly facilitate that conduct. Chiu's letters argue that continuing to distribute and profit from apps built for this purpose, after being put on notice, meets that facilitation standard.

"Apple and Google are profiting off apps that exploit women and girls by generating nonconsensual intimate deepfakes," Chiu said in a statement accompanying the letters. The 28 day window gives both companies a defined deadline to either comply or contest the legal theory, and Chiu's office has said it is prepared to seek civil penalties if the companies do not act.

How Did Apple and Google Respond?

Both companies say they have already acted on the specific apps named in Chiu's letter, though their public statements stop short of addressing why the apps were available in the first place. An Apple spokesperson said nudify apps are forbidden from the App Store under existing developer guidelines and that the company had "removed three of the apps in question and are in the process of terminating their developer accounts from our program." That leaves several of the eight apps Chiu's office flagged still unaccounted for in Apple's public response, with reporting indicating Apple remains in contact with the remaining developers over policy violations.

A Google spokesperson said all five Play Store apps referenced in the letter had been suspended, adding that "when violations are reported to us, we investigate and take swift action, which in the case of these apps has included suspending hundreds of violating apps and restricting related search terms like 'nudify' on our store." Google's statement frames this incident as part of an ongoing, larger enforcement effort rather than a one time response.

Why Did This Take a City Attorney's Letter?

Both Apple and Google have long banned nonconsensual sexual content from their app stores in writing. The gap between that policy and 13 apps circulating for enough time to draw a city attorney's formal investigation is the real story here. App store review processes are built to catch apps that describe prohibited functionality upfront, not tools marketed as innocuous face swap or photo editors that reveal their actual purpose only after a user downloads and opens them.

That gap matters because it shows automated and manual review alone did not catch these apps until outside pressure forced a second look. Chiu's 2025 lawsuit against 16 websites offering similar nonconsensual image tools suggests this is a pattern his office is tracking systematically, not an isolated case. For platforms that host third party apps at scale, the incident raises a harder question: whether policy enforcement can keep pace with generative AI tools that are cheap to build and easy to disguise. It is the same enforcement gap regulators are chasing with the EU's Digital Fairness Act crackdown on dark patterns, where platform design outpaces the rules meant to govern it.

What Happens Next?

Apple and Google now have until mid August 2026 to respond to Chiu's letters in full. If either company fails to satisfy the city attorney's office that the flagged apps, and any close variants, are gone for good, San Francisco has signaled it will pursue civil penalties under the state's deepfake facilitation statute. That would mark one of the first tests of whether a platform, rather than an app's individual developer, can be held civilly liable for hosting nonconsensual deepfake tools after being formally notified.

The broader question extends past these 13 apps. Generative AI has made building a nudify tool trivial compared to a few years ago, and disguising one as a generic photo editor is not difficult either. It echoes the identity verification debate playing out around Illinois HB 5511's push for device level age checks, another case where lawmakers are asking platforms to police content they did not build and cannot fully see inside of. Whether Apple and Google respond with faster proactive detection, or simply keep responding app by app to the next city attorney's letter, will determine how much of this enforcement burden continues to fall on victims and prosecutors rather than the platforms distributing the apps in the first place.

Looking Ahead

This case sits alongside a wider pattern of AI tools that impersonate real people without consent, from synthetic voices used in scam calls, as seen in the FBI's warnings about AI voice cloning fraud, to nonconsensual imagery generated at app store scale. Regulators are still working out who bears responsibility when a platform hosts the tool but does not build it. San Francisco's 28 day deadline will offer an early answer, and other city and state attorneys are likely watching how Apple and Google respond before deciding whether to send letters of their own.