What is a Deepfake, and How Are They Made?
A deepfake is a fabricated video that makes it possible to literally put words in someone’s mouth. As demonstrated by multiple research teams, while this technology is not yet perfect, it can be used to create a very convincing video or an even more convincing still shot.
Deepfakes can be made using a combination of techniques and tools.
Video Deepfakes
Using a specialized software solution, a video is scanned to identify the phenomes (the different sounds that make up full words) that are vocalized. Once they are identified, the phenomes are matched with the facial expressions that produce those sounds (also known as visemes). A 3D model of the subject’s face is then built based on the original video.
With the right software solution, these three factors can be combined with a transcript to create new footage and superimpose it over the original, making it appear that the person depicted is saying something that they never said. This creates a video that is just different enough to be disconcerting.
A similar method, that relies on mapping the expressions a person makes in source footage and applying them to a second person’s face, can even be used to bring paintings and old photographs to life.
https://youtu.be/p1b5aiTrGzY?t=257
Still Image Deepfakes
While the still images of people produced by AI were initially low-quality and generally unconvincing back in 2014, today’s are effectively indistinguishable from the real deal. This is thanks to a technique known as a generative adversarial network. Using this technique, one AI generates images of people’s faces, anticipating feedback on how it did. In order to reach the desired level of photo-realism, it could potentially take millions of repetitions.
Instead of subjecting a human being to the process of critiquing millions of images, a second AI is used to guess whether the picture was actually created by the first AI, or is a legitimate picture. While neither is particularly effective when first starting out, they swiftly improve in their capabilities over time, and can soon make images that are effectively indistinguishable from actual photographs of real people.
Neither of the people pictured here actually exist. They were instead created by NVIDIA in one of their machine learning AI initiatives.
As a result, we can see a dramatic rise in the capability of people to spread falsehoods and make the internet a misinformative place. Researchers at MIT Media Lab have spent years studying how well ordinary people can identify AI-manipulated media, and the results are sobering.
How Misinformative?
Let’s look at a relatively recent example of how impactful altered video can be. In May of 2019, a video of House Speaker Nancy Pelosi spread widely on social media, making her appear to be delivering a speech while intoxicated. The original footage had been slowed to 75 percent of its original speed, with the Speaker’s voice also adjusted to match her natural pitch. This was not an isolated incident, either. Similar manipulations of her voice appeared in separate videos from the same period.
Regardless of your political views, it cannot be denied that this kind of activity is dangerous. The Director of National Intelligence has previously alerted Congress to the possibility that America’s rivals could leverage deepfakes as part of future disinformation campaigns.
Consider this practical demonstration from a TED Talk by researcher Supasorn Suwajanakorn. Using footage of President George W. Bush, Suwajanakorn created maps of assorted public figures and had them appear to speak on his behalf. He also created four different models of President Barack Obama, all appearing to make an identical speech.
https://www.youtube.com/watch?v=o2DDU4g0PRo
In response to concerns about misuse, Suwajanakorn went on to work on a browser plugin called Reality Defender, designed to help natively identify deepfake videos and stop the spread of disinformation. Reality Defender has since grown into a full enterprise platform used by organizations to detect AI-manipulated media across video, audio, and images in real time.
How Could This Impact Your Business?
Sure, global politics and pop culture are significant to many people. But how much of a threat could deepfakes really be to your business? Unfortunately, more than you might expect.
The threat has already moved from theory to practice. In 2024, a finance employee at the global engineering firm Arup was convinced to transfer the equivalent of $25 million after scammers used AI-generated video to impersonate senior company executives during a video conference call. According to a Deloitte survey, nearly 26% of executives reported their company experienced at least one deepfake incident targeting financial or accounting data in a single 12-month period.
Think about what that means for your organization. A disgruntled former client or competitor could fabricate video of your executives making statements they never made. Your employees could be targeted by a scam that convincingly impersonates your CEO. These are not hypothetical scenarios anymore.
In addition to managed detection and response capabilities that help identify and contain threats quickly, there are several organizational steps businesses should take to reduce their exposure to deepfake-related fraud. Establishing clear out-of-band verification protocols for financial transfers, training employees to recognize the warning signs of synthetic media, and working with an IT partner experienced in modern security and compliance frameworks are all important layers of defense.
Beyond targeted fraud, there are plenty of other cybersecurity threats that can damage your business and its reputation just as seriously as any fabricated video.
Protecting Your Business in an Era of Deepfakes
Deepfake technology is not slowing down. The same AI advances that make these videos harder to create also make them harder to detect without the right tools and training in place. Businesses that treat this as a distant problem are leaving themselves exposed.
The most effective defense is a layered one: employee awareness, strong identity verification processes, and a trusted IT partner who stays ahead of emerging threats. WheelHouse IT helps organizations in South Florida, New York, and Los Angeles build the security posture they need to operate with confidence, regardless of how the threat landscape evolves.
Ready to talk about how to protect your business? Contact WheelHouse IT to start the conversation.
