Apple to certify authenticity of iPhone photos against AI

Apple is making another move in the war against visual disinformation. The Cupertino company is developing an internal tool called Apple Reference Image, detected in iOS 27 beta 5 by 9to5Mac. This feature aims to certify that a photograph was captured with an iPhone camera, using unique data linked to the sensor hardware. Although not yet active, its presence in the code suggests advanced development and a clear strategy by the company.

This move is part of the growing concern about AI-generated images that pass as real photographs, posing risks of disinformation, fraud, and manipulation of public opinion. According to a 2024 Europol report, 90% of fake visual content circulating online could be AI-generated by 2026, underscoring the urgency of verification solutions.

Why is it important?

The proliferation of photorealistic AI-generated images (such as those from ChatGPT or Gemini) has made it increasingly difficult to distinguish real from synthetic. So far, efforts have focused on adding watermarks to AI-generated images, as Google, Meta, and OpenAI do. However, these watermarks can be removed or may not be present in all tools. A study by the University of California, Berkeley, showed that current watermarks can be removed with a simple crop or compression, making them unreliable.

Apple proposes an inverse approach: instead of marking the fake, certify the authentic. By linking authenticity to camera hardware, a cryptographic guarantee is created that the image has not been manipulated since capture. This system is based on generating a unique key per device, similar to the Secure Enclave technology used in iPhones to protect biometric data. This key is combined with the photograph's metadata to create a digital seal impossible to forge without hardware access.

This approach could become a trust standard for photojournalism, legal documentation, and social media. For example, in the judicial sphere, photographs presented as evidence could be automatically verified, reducing the risk of manipulation. In journalism, news agencies could authenticate images received from reporters on the ground, guaranteeing their origin. Additionally, social platforms could display a "verified photo" badge on posts, similar to account verification.

Consequences and historical context

This development is not an isolated case. In 2023, Adobe launched the Content Authenticity Initiative (CAI) along with The New York Times and Twitter, adding provenance metadata to images. However, Apple's solution differs by operating at the operating system and hardware level, which could offer greater integration and ease of use. While CAI depends on collaboration from creators and platforms, Apple's approach is more proactive, as the device itself generates the proof of authenticity without requiring additional user actions.

Privacy is a key pillar: Apple has confirmed that photos will not be uploaded to its servers in their original form. The analysis will be performed on Private Cloud Compute, an infrastructure that processes data securely without storing it. This reinforces the company's privacy philosophy, something its users especially value. At a time when privacy has become a competitive differentiator, Apple strengthens its position against Google, which has been criticized for its data collection practices.

Historically, Apple has been a pioneer in privacy protection with features like App Tracking Transparency (2021) and On-Device Processing for Siri. This new tool aligns with that trajectory, offering a solution that does not compromise user privacy.

What should readers know?

• Availability: The feature is in development and could debut with iOS 27, expected to launch in September 2026, alongside new iPhones. According to 9to5Mac, the tool appears in beta 5, suggesting development is quite advanced.
• Compatibility: It is expected to work on devices with recent chips that include the necessary hardware to generate authentication keys. This likely includes iPhones with the A16 chip or later, and could extend to future iPhone 18 models, rumored to incorporate a new security engine.
• Limitations: Certification will only validate that the image comes from an iPhone, but will not protect against subsequent manipulations if metadata is removed or the image is recomposed. For example, if someone crops the image or edits it with Photoshop, the signature could be invalidated. However, Apple could implement a system that detects manipulations and marks them as "modified."
• Impact on the ecosystem: It could pressure Android and other manufacturers to adopt similar standards, and platforms like Instagram or X to display authenticity badges. Samsung has already shown interest in similar technologies, and Google is working on its own verification system for Android, as reported by The Verge in May 2026.

"Authenticity becomes a premium feature of the iPhone, differentiating it in a market where digital trust is increasingly valuable."

Analysis and speculation

It is important to note that, being leaked information from a beta, it is not officially confirmed by Apple. However, the direction is consistent with industry trends. Apple is likely to announce this feature at WWDC 2026, along with other iOS 27 news. The company typically presents its innovations at this event, held in June, to generate anticipation before the September launch.

There are open questions about how it will interact with the C2PA standard (Coalition for Content Provenance and Authenticity), of which Apple is a member. It could integrate to interoperate with other platforms, allowing Apple-verified images to be recognized by other applications and services. This would be key for the solution to have a global impact. According to The Information, Apple is already in talks with Adobe and Microsoft to define common standards.

In summary, Apple Reference Image is a significant step toward content verification, with implications for digital trust and the fight against visual disinformation. If Apple successfully implements it, it could establish a new gold standard in image authentication, pressuring the entire industry to follow suit. Digital trust is becoming a key battleground, and Apple seems determined to lead this race.