This article originally appeared in Ecommerce Times on June 27, 2019.
The U.S. House Intelligence Committee last week heard expert testimony on the growing threat posed by “deepfakes” — altered videos and other artificial intelligence-generated false information — and what it could mean for the 2020 general elections, as well as the country’s national security overall.
The technologies collectively known as “deepfakes” can be used to combine or superimpose existing images and videos with other images or videos by utilizing AI or machine learning “generative adversarial network” techniques.
These capabilities have allowed the creation of fake celebrity videos — including pornography — as well as for the distribution of fake news and other malicious hoaxes.
The hearing followed the widespread online distribution of a doctored video of House Speaker Nancy Pelosi, D-Calif., which made her appear impaired. The video made the rounds on social media and was viewed more than 2.5 million times on Facebook.
Deepfakes has become a bipartisan issue, with both Democrats and Republicans expressing concerns over the use of manipulated videos as a tool of disinformation.
The House Intelligence Committee heard testimony from four different experts in AI and disinformation about the potential risks to the U.S. government and even democracy from deepfakes. However, one expert also warned of the threat deepfakes could pose to the private sector. One such scenario might involve a deepfake video showing a CEO committing a crime. Putting that type of video in circulation could impact a company’s stock price.
Whether in politics or the business world, even if a video is debunked the damage could be lasting.
The phrase “deepfakes” was first coined in 2017, but the ability to modify and manipulate videos goes back to the Video Rewrite program, which was published in 1997. It allowed users to modify video footage of a person speaking to depict that person mouthing the words from a completely different audio track.
The technique of combining videos and changing what was said has been used in Hollywood even longer, but it generally was a costly and time-consuming endeavor. The film Forrest Gump, for example, required a team of artists to render the character, played by Tom Hanks, into historic footage. Now, more than 20 years later, the results aren’t nearly as good as what today’s software can do.
Simple programs such as FakeApp — which was released in January 2018 — allow users to manipulate videos easily, swapping faces. The app utilizes an artificial neural network and just 4 GB of storage to generate the videos.
The quality and detail of the videos is based on the amount of visual material that can be provided, but given that today’s political figures appear in hundreds of hours of footage, it is easy enough to make a compelling video.
Fighting the Fakes
Technology to combat deepfakes is in development. The USC Information Sciences Institute developed a tool that can detect fakes with up to 96 percent accuracy. It is able to detect subtle face and head movements, as well as unique video “artifacts” — the noticeable distortion of media that is caused by compression, which also can indicate a video manipulation.
Previous methods for detecting deepfakes required frame-by-frame analysis of the video, but the USC ISC researchers developed a tool that has been tested on more than 1,000 videos and has proven to be less computationally intensive.
It could have the potential to scale and be used to automatically — and more importantly quickly — detect fakes as the videos are uploaded on Facebook and other social media platforms. This could allow near real-time detection, something that would keep such videos from going viral.
The USC ISI researchers rely on a two-step process. It first requires that hundreds of examples of verified videos of a person are uploaded. A deep learning algorithm known as a “convolutional neural network” then allows researchers to identify features and patterns in an individual’s face. The tools then can determine if a video has been manipulated by comparing the motions and facial features.
The results are similar to a biometric reader that can recognize a face, retina scan or fingerprint — but just as with those technologies, a baseline is required for comparison. That could be easy for famous individuals such as Speaker Pelosi or actor Tom Hanks, but for the average person it probably won’t be as easy, as the database of existing video footage may be limited or nonexistent.
Potential to Be Weaponized
Deepfakes have the potential to be far worse and do far more damage than “Photoshopped” images — both on an individual and even national level.
“There is a world of difference between Photoshopped images and AI-aided videos, and people should be concerned with deepfakes because of their heightened realism and potential for weaponization,” warned Usman Rahim, digital security and operations manager for The Media Trust.
One reason is that today people accept that photos can be altered, so much so that these have earned the moniker “cheapfakes.” Video is a new frontier.
“Much fewer are aware of how realistic fake videos have become and how easily they can be made in order to spread disinformation, destroy reputations, or disrupt democratic processes,” Rahim told the E-Commerce Times
“In the wrong hands, deepfakes spread through the Internet, especially social media, can have a large impact on individuals — and more broadly, societies and economies,” he added.
“Aside from the National security risk — e.g., a deep fake video of a world leader used to incite terrorist activity — the political risk is especially high in a competitive national election such as 2020, with multiple candidates seeking to unseat a controversial incumbent,” noted associate professor Larry Parnell, strategic public relations program director in The Graduate School of Political Management at George Washington University.
“Either side might be tempted to engage in this activity, and that would make ‘old school’ dirty tricks seem mundane and quaint,” he told the E-Commerce Times. “We have already seen how social media can be used to impact a national election in 2016. That will seem like child’s play compared to how advanced this technology has become in the last two-to-three years.”
Beyond Politics and Security Risks
Deepfakes could present a problem on a much more personal and individual level. The technology already has been used to create revenge porn videos, and the potential is there to use it for other sinister or nefarious purposes.
“In the hands of unsupervised kids, deepfakes can raise cyberbullying to a new level,” said The Media Trust’s Rahim.
“Imagine what happens if our own or our children’s images are used and distributed online,” he added.
“We might even see fake videos and social media posts being used in legal proceedings as evidence against a controversial figure to silence them or destroy their credibility,” warned GW’s Parnell.
There already have been calls to hold the tech industry responsible for the creation of deepfakes.
“If you create software that allows a user to create deepfakes, well, then you will be held liable for significant damages, maybe even held criminally liable,” argued Anirudh Ruhil, a professor in the Voinovich School of Leadership and Public Affairs at Ohio University.
“Should you be a social media or other tech platform that disseminates deepfakes, you will be held liable and pay damages, maybe even jail time,” he told the E-Commerce Times.
“These are your only policy options, because otherwise you will have the social media platforms and websites going scot-free for pushing deep fakes to the mass public,” Ruhil added.
It is possible the authors of such heinous videos may not be found easily, and in some cases could be a world way, making prosecution a non-starter.
“In some ways, this policy is similar to what someone might argue about gun control: Target the sellers of weapons capable of causing massive damage,” explained Ruhil. “If we allow the tech industry to skate free, you will see repeats of the same struggles we have had policing Facebook, YouTube, Twitter and the like.”
The good news about deepfakes is that in many cases the technology still isn’t perfect, and there are plenty of telltale signs that the video has been manipulated.
Also, there are already tools that can help researchers and the media tell fact from fiction.
“Social media and platforms, and traditional media can use these tools to identify deepfakes and either remove them or label them as such, so users aren’t fooled,” said Rahim.
Another solution could be as simple as adding “digital noise” to images and files, making it harder to use them to produce deepfakes.
However, just as in the world of cybersecurity, it’s likely the bad actors will stay one step ahead — so today’s solutions may not solve tomorrow’s methods for producing deepfakes.
It may be necessary to put more effort into solving this problem before it becomes so great that it is not solvable.
“While it may be a constant and expensive process — the major tech companies should invest now in emerging technology to spot deep fake videos,” suggested Parnell.
“Software is being developed by DARPA and other government and private sector companies that could be utilized, as the alternative is to be caught flat-footed and be publicly criticized for not doing so — and suffer the serious reputation damage that will result,” he added.
For now the best thing that can happen is for publishers and social media platforms to call out and root out deepfakes, which will help restore trust.
“If they don’t, their credibility will continue to dive, and they will have a hand in their own business’ demise,” said Rahim.
“Distrust of social media platforms in particular is rising, and they are seen almost as much of a threat as hackers,” he warned.
“The era of prioritizing the monetization of consumer data at the expense of maintaining or regaining consumer trust is giving way to a new era where online trust works hand-in-glove with growing your bottom line,” Rahim pointed out. “Social and traditional media can also be a force for good by outing bad actors and raising consumers’ awareness of the prevalence and threats of deepfakes.”