Throw a stone and also you’ll possible hit a deepfake. The commoditization of generative AI has led to an absolute explosion of faux content material on-line: According to ID verification platform Sumsub, there’s been a 4x improve in deepfakes worldwide from 2023 to 2024. In 2024, deepfakes accounted for 7% of all fraud, per Sumsub, starting from impersonations and account takeovers to classy social engineering campaigns.
In what it hopes will likely be a significant contribution to the combat in opposition to deepfakes, Meta is releasing a device to use imperceptible watermarks to AI-generated video clips. Announced on Thursday, the device, referred to as Meta Video Seal, is on the market in open supply and designed to be built-in into present software program. The device joins Meta’s different watermarking instruments, Watermark Anything (re-released right this moment beneath a permissive license) and Audio Seal.
“We developed Video Seal to supply a more practical video watermarking answer, significantly for detecting AI-generated movies and defending originality,” Pierre Fernandez, AI analysis scientist at Meta, informed TechCrunch in an interview.
Video Seal isn’t the primary expertise of its sort. DeepMind’s SynthID can watermark movies, and Microsoft has its personal video watermarking methodologies.
But Fernandez asserts that many present approaches fall quick.
“While different watermarking instruments exist, they don’t supply enough robustness to video compression, which could be very prevalent when sharing content material by means of social platforms; weren’t environment friendly sufficient to run at scale; weren’t open or reproducible; or had been derived from picture watermarking, which is suboptimal for movies,” Fernandez mentioned.
In addition to a watermark, Video Seal can add a hidden message to movies that may later be uncovered to find out their origins. Meta claims that Video Seal is resilient in opposition to frequent edits like blurring and cropping, in addition to in style compression algorithms.
Fernandez admits that Video Seal has sure limitations, primarily the trade-off between how perceptible the device’s watermarks are and their general resilience to manipulation. Heavy compression and important edits could alter the watermarks or render them unrecoverable, he added.
Of course, the larger downside dealing with Video Seal is that devs and business gained’t have a lot purpose to undertake it, significantly these already utilizing proprietary options. In a bid to deal with that, Meta is launching a public leaderboard, Meta Omni Seal Bench, devoted to evaluating the efficiency of assorted watermarking strategies, and organizing a workshop on watermarking this yr at ICLR, a serious AI convention.
“We hope that increasingly more AI researchers and builders will combine some type of watermarking into their work,” Fernandez mentioned. “We need to collaborate with the business and the tutorial neighborhood to progress sooner within the area.”
