Company to tag videos and images on Facebook and Instagram from May
Meta, the parent company of Facebook and Instagram, announced on Friday its plans to implement labeling for AI-generated content starting this May as it tries to reassure users and governments over the risks of deepfakes
The company aims to address concerns raised by users and governments regarding the risks associated with deepfakes. Instead of outright removing manipulated images and audio that do not violate its policies, the social media giant will rely on labeling and contextualization to provide transparency without unnecessarily restricting freedom of expression
Reports from tech insiders, such as The Verge, in February indicated that Meta would introduce labeling for AI-generated photos uploaded to Facebook, Instagram, and Threads, particularly as elections are being observed worldwide. Nick Clegg, Meta’s president of global affairs, highlighted in an interview the increasing difficulty in distinguishing machine-generated content from reality. Meanwhile, the White House urged companies to watermark AI-generated images and videos, a measure Meta is actively developing tools to implement.
Meta has already begun adding a watermark stating “Imagined with AI” to images generated using its Imagine AI generator and plans to extend this feature to various tools, including those from Google, OpenAI, Microsoft, and Midjourney, among others.
In a blog post from February 2024, Meta urged people to consider several factors when evaluating AI-generated content, including the trustworthiness of the account and the unnaturalness of the content.