Universal Detectors Identify AI Deepfake Videos with Unprecedented Accuracy

Deepfake video showcasing Australian Prime Minister Anthony Albanese on a smartphone

Australia’s Associated Press/Alamy

Universal DeepFake Detectors have demonstrated optimal accuracy in identifying various types of videos that have been altered or entirely produced by AI. This technology can assist in flagging adult content, deepfake scams, or misleading political videos generated by unregulated AI.

The rise of accessible DeepFake Creation Tools powered by inexpensive AI has led to rampant online distribution of synthetic videos. Numerous instances involve non-consensual depictions of women, including celebrities and students. Additionally, deepfakes are utilized to sway political elections and escalate financial scams targeting everyday consumers and corporate leaders.

Nevertheless, most AI models designed to spot synthetic videos primarily focus on facial recognition. This means they excel in identifying a specific type of deepfake where a person’s face is swapped with existing footage. “We need a single video with a manipulated face and a model capable of detecting background alterations or entirely synthetic videos,” states Rohit Kundu from the University of California Riverside. “Our approach tackles that particular issue, considering the entire video could be entirely synthetically produced.”

Kundu and his team have developed a universal detector that leverages AI to analyze both facial features and various background elements within the video. It can detect subtle signs of spatial and temporal inconsistencies in deepfake content. Consequently, it identifies irregular lighting conditions for people inserted into face-swapped videos, as well as discrepancies in background details of fully AI-generated videos. The detector can even recognize AI manipulation in synthetic videos devoid of human faces, and it flags realistic scenes in video games like Grand Theft Auto V, independent of AI generation.

“Most traditional methods focus on AI-generated facial videos, such as face swaps and lip-synced content.” says Siwei Lyu from Buffalo University in New York. “This new method is broader in its applications.”

The universal detector reached an impressive accuracy rate of 95% to 99% in recognizing four sets of test videos featuring manipulated faces. This performance surpasses all previously published methods for detecting this type of deepfake. In evaluations of fully synthetic videos, it yielded more precise results than any other detectors assessed to date. Researcher I presented their findings at the 2025 IEEE Conference on Computer Vision and Pattern Recognition in Nashville, Tennessee, on June 15th.

Several researchers from Google also contributed to the development of these new detectors. Though Google has not responded to inquiries regarding whether this detection method would be beneficial for identifying deepfakes on platforms like YouTube, the company is among those advocating for watermarking tools that help label AI-generated content.

The universal detectors have room for future enhancements. For instance, it would be advantageous to develop capabilities for detecting deepfakes utilized during live video conference calls—a tactic some scammers are now employing.

“How can you tell if the individual on the other end is genuine or a deepfake-generated video, even with network factors like bandwidth affecting the transmission?” asks Amit Roy-Chowdhury from the University of California Riverside. “This is a different area we’re exploring in our lab.”

Topics:

Source: www.newscientist.com

The Mystery of Missile Detectors and Santa Trackers at the Festival

early christmas

Today is my vacation. Actually, it's not a vacation. Because of printing schedules, Feedback is writing this festive edition in the first week of December. We ordered a gift for Mrs. Feedback, but we still don't know what Feedback Jr. will get Mrs. Feedback (with Feedback's money). There is no decoration at all. And we're staring down the barrel of multiple school events. We will be asked to dress up in a festive mood for our children. In short, the feedback is moody.

So now seems like a good time to complain NORAD tracks Santaone of those efforts where you can't believe the feedback exists at all. NORAD, of course, is the North American Aerospace Defense Command, and its primary mission is to use a combination of satellites and radar to detect things flying into North American airspace, such as Chinese weather balloons. But on December 24th, an elaborate volunteer-run exercise to track Santa as he travels around the world will take place. You can also call for updates, check out his progress on his website, and follow him on social media.

How this tradition began is interestingly complex. A fact sheet on the NORAD website states: heartwarming story: “NORAD has been tracking Santa since 1955, when a young child accidentally dialed an unlisted telephone number for the Continental Air Defense Command (CONAD). [the precursor to NORAD] Operations Center…She believed she called Santa Claus after seeing an advertisement in the local newspaper. Air Force Col. Harry Shoup, the commander on duty that night, quickly realized the mistake and promised that CONAD would ensure Santa's safe journey from the North Pole. ”

But this story truth?This is difficult to fully elucidate. Details vary by account. How Shoup first reactedto how many times to call I came that first year. However, what is clear is that All this It started at the height of the Cold War. NORAD is inherently scary because it exists to detect incoming nuclear missiles. Its headquarters is literally a bunker dug into the mountain. stunt track santa This was and still is a great way to look cute. On the other hand, if a nuclear war were to begin, NORAD would almost certainly play a key role in it. On the other hand, run the cute Santa Tracker.

But the biggest gripe with feedback on the whole thing is that I'm 90 percent sure this stunt is the inspiration for a new Christmas-themed fantasy-action-comedy-thriller. red one. In the film's stacked qualifiers, Santa's journey is reimagined as a militaristic security operation involving a fighter jet escort and a sort of Arctic secret service commanded by Dwayne Johnson. It looks like one of the worst movies of the year, and I strongly suspect that the feedback is ultimately NORAD Tracks Santa's fault.

long setup

Speaking of tracking immaterial things, an Australian research project is asking people to monitor their own flatulence. It is necessarily Graph a fart.

Using a free phone app, participants cantrack Quantity and quality of output, including attributes such as odor, volume, duration, aftertaste, and detectability. ” feedback appreciates the precisely guided use of the word “residual” and points out that fart residual can be scored from “momentary” to “permanent.”

It aims to further explain “one of the main gut health symptoms experienced by Australians'', specifically “excessive bloating'', which up to 43 per cent of Australians report experiencing almost every day. It's about deepening your understanding.

Feedback is not supported guardianThe heading for the project is “wind power”. We also cannot approve of the decision to limit the study to Australia. But thankfully, others will too, as shown in this amazing 42-page feature on researchers trying to catch farts. And their work is extremely important at this time, when millions of us, in many countries, eat appalling amounts of meat, mince pies, and above all, gas. I'm about to eat some Brussels sprouts.

cheap in price

For all our readers who don't have a present for their loved ones yet, you're out of luck. I missed a great opportunity. London auction house Christie's recently announced the first-ever science fiction and fantasy auctionthe highlights are: dune bible: A circa 1975 collection of behind-the-scenes documents from Alejandro Jodorowsky's unproduced film adaptation of Frank Herbert's book. dunes. This is an ideal excuse for feedback to bring something to our chest. So here we go.

Jodorowsky's dunes The film has an almost mythical status as one of the great assumptions of science fiction cinema. The blockbuster, starring Orson Welles and Mick Jagger, with production design by H.R. Giger (later a director), is expected to be at least 10 hours long. alien fame). If he had succeeded in it, it would have become a classic.

Here's the problem. Jodorowsky is one of the most exaggerated filmmakers to have written a screenplay that included depictions of his own penis. Feedback learned from Danny Peary cult moviecovering Jodorowsky's mysterious 1970 Western. el topo. Peary complained that there were “too many references, Jungian and religious symbols/artifacts…inside jokes, and too many vague images that no one but Jodorowsky knows what's going on.” . Imagine it lasting more than 10 hours. we are saved.

Have a story for feedback?

You can email your article to Feedback at feedback@newscientist.com. Please enter your home address. This week's and past feedback can be found on our website.

Source: www.newscientist.com