Skip to main content
MagazineCoverage

Google's latest technology can distinguish between real, edited, and AI images

A system that can differentiate between real, altered, and AI-generated photographs will shortly be released by Google.

2 min read0 views
Google's latest technology can distinguish between real, edited, and AI images
Sharefin

A system that can differentiate between real, altered, and AI-generated photographs will shortly be released by Google. The updated "About this image" function in search results can now show whether an image was altered by artificial intelligence.

Google's system records an image's digital history across hardware and software and is a component of the Coalition for Content Provenance and Authenticity (C2PA).

This initiative is aimed at improving transparency and fostering user trust as AI-generated media becomes more prevalent. The new labeling system will be integrated with C2PA metadata, which provides information about the origin and modification history of digital content.

By using C2PA metadata, Google aims to give users more context about the images they encounter, making it easier to identify the authenticity and provenance of visual content online.

While Google's adoption of the C2PA standard marks a significant move toward combating the spread of AI-generated imagery and ensuring content authenticity, widespread adoption and interoperability remain key challenges.

Presently, only a limited number of cameras from Leica and Sony support the C2PA standard, highlighting the need for broader industry participation. For the initiative to be truly effective, major camera manufacturers like Nikon, Canon, Apple, and Google need to incorporate this standard into their devices.

Moving forward, this would enable a more consistent and reliable method for verifying the authenticity and origin of digital images across different platforms and devices, ultimately bolstering efforts to curb misinformation and enhance transparency in digital media.