Meta will detect and label images generated by generative artificial intelligence from other companies, such as Google and OpenAI, but that have been published on their platforms. The information was revealed by Nick Clegg, president of global affairs at big tech, in an article on Meta’s official blog on Tuesday (6).
Meta will use a set of invisible markers present in the files to identify AI-generated images. Thus, content published on Facebook, Instagram and Threads will have tags informing them that they are digital creations.
Meta says the label system for AI-created images aims to inform people that certain realistic content is digital creations. It is worth mentioning that big tech already uses tags to flag images generated with its own AI tools, such as Meta AI Imagine.
The new system, expected to be deployed in the coming months, will label content created on AI platforms from big tech companies. These include OpenAI, Microsoft, Adobe, Midjourney, Shutterstock, and Google.
Meta’s announcement reveals an early look at creating a standard system for identifying content made with generative AI. According to experts, this will be a necessary tool to avoid possible damage caused by realistic materials shared on the internet.
Previously, Nick Clegg told Reuters he was confident with Meta’s solution for labeling AI-generated images. The executive also stated that tools to identify audio and video created with AI were under development, but were “more complicated” techniques.
WhatsApp Could Get Free From AI Labels
According to Reuters, Meta’s spokesperson declined to say whether the label system for AI-made content would also be used on WhatsApp. Remembering that, unfortunately, the big tech messenger application is also used by malicious users to spread fake news and images.
Recently, Meta’s independent oversight board took a stand against the policy of deleting videos doctored with misleading content. The representatives claim that the materials should be labeled rather than removed from big tech platforms.
Nick Clegg agrees with the group’s decision and cites that the initial policy adopted by the company was not adequate. So people are going to have to adapt to “an environment where you’re going to have a lot more synthetic and hybrid content than before.”
