Meta updates AI detection label from "Made with AI" to "AI Info"

New Delhi: Meta updated its AI content detection label from "Made with AI" to "AI Info" on Monday, following criticism from Instagram influencers and photographers who found their original posts inaccurately labeled as AI-generated. 

The company acknowledged that the previous label did not meet user expectations and often lacked sufficient context. The new label aims to reduce confusion by providing clearer information about images and videos edited with AI tools.

In an updated announcement, Meta clarified its approach to labeling AI-generated content, emphasizing the transition from "Made with AI" to "AI Info." 

Despite this change, Meta will continue to mark posts with any AI-assisted edits, no matter how minor. These updated labels are expected to roll out to all users in the coming days.

Users can now click on the "AI Info" label to access more detailed information. A bottom sheet will explain that generative AI may have been used in creating or editing the content, along with an introduction to generative AI for users unfamiliar with the technology.

Meta employs industry-standard indicators, such as those from the Coalition for Content Provenance and Authenticity (C2PA) and the International Press Telecommunications Council (IPTC), to detect AI-generated content. However, challenges remain as many AI-generated images still evade detection by these tools.

There is some concern that labeling even AI-assisted retouched images with "AI Info" might dilute the label's original purpose of identifying deepfakes and manipulative AI content. 

Meta intends to collaborate with industry partners to refine its AI labeling process to better align with user expectations and improve detection accuracy.