Facebook to drop Face Recognition, will delete templates on more than one billion people
Facebook is dropping its Face Recognition tagging feature and will delete templates on more than one billion people, the service announced in a blog post Tuesday. The social media network “will delete more than a billion people’s individual facial recognition templates,” wrote Jerome Pesenti, vice president of artificial intelligence at Meta.
(Meta is a rebrand of Facebook’s parent company announced last month amidst reporting on internal Facebook documents that show the company was aware of harmful outcomes of its platform, including amplification of radicalizing misinformation and negative mental health outcomes for teen users. Company founder Mark Zuckerberg announced the name change at the same time as plans to double down on building the Metaverse, an immersive virtual experience named after the term for such worlds coined in a dystopian Neal Stephenson science fiction novel.)
Facebook has long used its access to a vast cache of visual data to power features that can automatically recognize people in photos and videos. It’s also faced legal challenges over the privacy implications of those features—including a class action suit under Illinois’ biometric privacy law.
The facial recognition change was met by surprise and delight by privacy advocates and researchers.
Facebook has announced they will be deleting over a billion face recognition templates as they shut down their entire face recognition system. This is great news for Facebook users, and for the global movement pushing back on this technology. https://t.co/0ErdCBhkCT
— EFF (@EFF) November 2, 2021
Other companies have previously paused some facial recognition efforts amidst public backlash. In May, Amazon announced it would indefinitely extend a ban on police use of its facial recognition systems. Amazon first announced the prohibition during the racial justice protests during the summer of 2020 and after pitching its facial recognition products to U.S. Immigration and Customs Enforcement as the Trump administration cracked down on immigration.
Research shows facial recognition technology is less accurate at identifying people with darker skins—and such tools are already connected to reports of Black people being arrested and imprisoned due to misidentification.
Pesenti wrote the decision to drop Face Recognition from Facebook was driven by larger societal concerns and a lack of clear regulatory guidance. While dropping this major implementation of facial recognition technology, the company will continue to explore its use in more limited use cases—including to unlock accounts, for example, he wrote.
The decision to drop the Face Recognition feature is part of “a company-wide move away from this kind of broad identification, and toward narrower forms of personal authentication,” according to the blog post.
Andrea Peterson
(they/them) is a longtime cybersecurity journalist who cut their teeth covering technology policy at ThinkProgress (RIP) and The Washington Post before doing deep-dive public records investigations at the Project on Government Oversight and American Oversight.