Deepfake-hunting A.I. could help strike back against the threat of fake news

Sylvester Stallone deepfake of him starring in Terminator 2: Judgement Day
Sylvester Stallone deepfake (replacing Arnold Schwarzenegger in Terminator 2: Judgement Day) Ctrl Shift Face/Youtube

Of all the A.I. tools to have emerged in recent years, very few have generated as much concern as deepfakes. A combination of “deep learning” and “fake,” deepfake technology allows anyone to create images or videos in which photographic elements are convincingly superimposed onto other pictures. While some of the ways this tech has been showcased have been for entertainment (think superimposing Sylvester Stallone’s face onto Arnie’s body in Terminator 2), other use-cases have been more alarming. Deepfakes make possible everything from traumatizing and reputation-ruining “revenge porn” to misleading fake news.

As a result, while a growing number of researchers have been working to make deepfake technology more realistic, others have been searching for ways to help us better distinguish between images and videos which are real and those that have been algorithmically doctored.

At Drexel University, a team of researchers in the Multimedia and Information Security Lab have developed a deep neural network which can spot manipulated images with a high degree of accuracy. In the process, its creators hope that they can provide the means of fighting back against the dangers of deepfakes. It’s not the first time researchers have attempted to solve this problem, but it is potentially one of the most promising efforts to materialize so far in this ongoing cat-and-mouse game.

“Many [previous] deepfake detectors rely on visual quirks in the faked video, like inconsistent lip movement or weird head pose,” Brian Hosler, a researcher on the project, told Digital Trends. “However, researchers are getting better and better at ironing out these visual cues when creating deepfakes. Our system uses statistical correlations in the pixels of a video to identify the camera that captured it. A deepfake video is unlikely to have the same statistical correlations in the fake part of the video as in the real part, and this inconsistency could be used to detect fake content.”

The project started out as an experiment to see whether it was possible to create an A.I. algorithm able to spot to difference between videos captured by different cameras. Like a watermark, every camera captures and compresses videos slightly differently. Most of us can’t do it, but an algorithm that’s trained to detect these differences can recognize the unique visual fingerprints associated with different cameras and use this to identify the format of a particular video. The system could also be used for other things, such as developing algorithms to detect videos with deleted frames, or to detect whether or not a video was uploaded to social media.

How they did it

The Drexel team curated a large database of videos, running to around 20 hours, from 46 different cameras. They then trained a neural network to be able to distinguish these elements. As convincing as a deepfake video may look to your average person, the A.I. examines them pixel by pixel to search for elements which have been altered. Not only is the resultant A.I. able to recognize which pictures had been changed, it is also able to identify the specific part of the image that has been doctored.

Owen Mayer, a member of the research group, has previously created a system which analyzes some of these statistical correlations to determine if two parts of an image are edited in different ways. A demo of this system is available online. However, this latest work is the first time that such an approach has been conducted in video footage. This is a bigger problem, and one which is crucial to get right as deepfakes become more prevalent.

Kim Kardashian Deepfake Interview Image
Kim Kardashian Deepfake Interview Image

“We plan to release a version of our code, or even an application, to the public so that anyone can take a video, and try to identify the camera model of origin,” Hosler continued. “The tools we make, and that researchers in our field make, are often open-source and freely distributed.”

There’s still more work to be done, though. Deepfakes are only getting better, which means that researchers on the other side of the fence must not rest on their laurels. Tools will need to continue to evolve to make sure that they can continue to spot faked images and video as they dispense with the more noticeable visual traits that can mark out current deepfakes as, well, fakes. As audio deepfake tools, capable of mimicking voices, continue to develop, it will also be necessary to create audio-spotting tools to track them down.

For now, perhaps the biggest challenge is to raise awareness of the issue. Like fact-checking something we read online, the availability of information on the internet only works to our advantage if we know enough to second-guess whatever we read. Until now, finding video proof that something happened was enough to convince many of us that it actually took place. That mindset is going to have to change.

“I think one of the largest hurdles to getting everyone to use forensic tools like these is the knowledge gap,” Hosler said. “We as researchers should make not only the tools, but the underlying ideas, more palatable to the public if we truly have an impact.”

Whatever form these tools take — whether it’s as a web browser plug-in or an A.I. that’s automatically employed by internet giants to flag content before it’s shown to users — we sure hope the right approach is employed to make these as accessible as possible.

Hey, it’s only the future of truth as we know it that’s at stake…

Emerging Tech

Amazing app promises a full fitness checkup from a 30-second selfie

Researchers at the University of Toronto have developed an app that's able to gather vital health information about users with nothing more invasive than a 30-second selfie. Here's how it works.
Emerging Tech

Astro the dog-inspired quadruped robot can sit, lie down, and… learn?

Move over Spot! Researchers from Florida Atlantic University have built a new dog robot called Astro. Thanks to deep learning technology, it promises to be able to learn just like a real dog.
Emerging Tech

Google’s soccer-playing A.I. hopes to master the world’s most popular sport

Think the player A.I. in FIFA ‘19 was something special? You haven’t seen anything yet! That’s because Google is developing its own soccer-playing artificial intelligence. And, if the company’s history with machine intelligence is…
Emerging Tech

The best deepfakes on the web: Baby Elon, Ryan Reynolds Wonka, and beyond

Deepfakes, the A.I.-aided face-swapping technology that threatens the future of truth as we know it, are everywhere. Here are some of the scariest, funniest, and most convincing we've seen.
Health & Fitness

We spit in a ton of test tubes to find the best and most unique DNA tests

DNA tests aren’t just limited to ancestry. You can test for your risks for certain diseases, the best workouts and diets for your health and fitness, and more.
Emerging Tech

Artificial tree promises to suck up as much air pollution as a small forest

Startup Biomitech has developed an artificial tree that it claims is capable of sucking up as much air pollution as 368 real trees. It could be a game-changer for cities with limited free space.
Emerging Tech

Awesome Tech You Can’t Buy Yet: Racing drones and robotic ping pong trainers

Check out our roundup of the best new crowdfunding projects and product announcements that hit the web this week. You may not be able to buy this stuff yet, but it sure is fun to gawk!
Emerging Tech

Mars 2020 rover now has a rotating array of drill bits for sampling Martian rock

Most the key components in the Mars 2020 rover are installed and ready to go. The next phase of construction was to install the bit carousel, an important mechanism for the gathering and sorting of samples from the Martian surface.
Emerging Tech

NASA selects landing site candidates for OSIRIS-Rex to sample asteroid Bennu

Last year, the OSIRIS-REx craft arrived at asteroid Bennu, from which it will collect a sample from the asteroid to be brought back to Earth. Now, the NASA team has selected four potential sites to choose from for the sampling mission.
Emerging Tech

NASA wants to send two more missions to Mars to collect rock samples

With its Mars 2020 mission, NASA hopes to collect samples from the surface of the planet. The challenge is how to get those samples back to Earth. Now, NASA has revealed its plans for two followup missions to Mars.
Emerging Tech

Eric Geusz: Apple engineer by day, spaceship designer by night

An Apple software engineer by day, artist Eric Geusz spends his nights drawing everyday household objects as amazing, science fiction-style spaceships. Check out the impressive results.
Emerging Tech

The black hole at the center of our galaxy is flaring and no one knows why

At the heart of our galaxy lies a supermassive black hole, Sagittarius A*. Normally this giant monster is relatively docile, but recently it's been a hotbed of unexpected activity, rapidly glowing 75 times brighter than normal.
Emerging Tech

SpaceIL’s crashed lander may have sent thousands of tardigrades to the moon

When the SpaceIL craft Beresheet crashed into the moon earlier this year, it left more than just an impact mark. Thousands of micro-animals called tardigrades were along for the ride and may have survived the crash.
Emerging Tech

NASA’s satellite projects will study the sun using solar sailing

Small satellites can be used for all sorts of purposes, and NASA has been searching for ideas to push ahead the capabilities of the hardware. The agency has announced two new projects to demonstrate the potential of small satellites.