Meta is changing AI tags after real photos were tagged as AI

Meta is changing the tags it applies to social media posts it suspects were generated in some way with artificial intelligence tools. The parent company of Facebook, Instagram, Threads and WhatsApp said its new tagline will feature “Info AI” alongside a post where it used to say “Made with AI”.

It’s making these changes in part because Meta’s detection systems were labeling images with minor edits as “Made with AI,” leading some artists to criticize the approach.

In one high-profile example, former White House photographer Pete Souza told TechCrunch that cropping tools appear to be adding information to images, and that the information was then alerting Meta’s AI detectors.

Meta, for its part, said it is striking a balance between fast technology and its responsibility to help people understand what its systems show in their feeds.

“As we work with companies across the industry to improve the process so that our labeling approach better aligns with our purpose, we’re updating the ‘Made with AI’ label to ‘Info AI’ across our apps, which people can click for more information,” the company said in a statement on Monday.

Read more: How close is that picture to the truth? What you need to know in the age of AI

Meta’s shifty approach underscores the speed at which AI technologies are spreading across the web, making it harder for ordinary people to discern what’s really real anymore.

This is particularly concerning as we head into the 2024 US presidential election in November, when people acting in bad faith are expected to increase their efforts to spread misinformation and ultimately confuse voters. Google researchers published a report last month underscoring this point, with the Financial Times reporting that AI creation of politicians and celebrities is by far the most popular use of this technology by bad actors.

Tech companies have tried to respond to the threat publicly. OpenAI earlier this year said it had disrupted social media disinformation campaigns linked to Russia, China, Iran and Israel, which were each being powered by its AI tools. Apple, meanwhile, announced last month that it will add metadata to label images, regardless of whether they are altered, edited or generated by AI.

However, technology seems to be moving much faster than companies’ ability to identify it. A new term, “slop,” has become increasingly popular to describe the growing flood of AI-generated posts.

Meanwhile, tech companies including Google have contributed to the problem with new technologies like its AI Roundup for Search, which was caught spreading racist conspiracy theories and dangerous health advice, including adding glue to pizza to prevent the cheese from slipping. Google, for its part, has since said it will slow down its rollout of AI Overviews, though some publications still found it recommending sticky pizza additives weeks later.