What's That Term?: Deep Fakes
Have you ever come across a video on social media that seems sort of "off" in some way and has something weird about the facial expressions that you just can't put your finger on. You might think your eyes are deceiving you, but it could actually be a deep fake instead.
What is it?
Deep fakes are a type of video manipulation where artificial intelligence is used to create a fake video of someone saying or doing something they didn't. They are made by combining a real video of someone with a deep learning algorithm that can generate a new face that perfectly matches the person in the original video. Over time, these types of videos and AI programs have gotten better at masking the new face and speech used. By now, it can be extremely hard to detect deep fakes, even for those savvy with the technology and the internet in general. Now just imagine those who have no idea the technology even exists! This was probably the case for those who first saw deep fakes used back in the late 1800s. While it was different from what we know today, the story is still reflective of how people make deep fakes
In the late 1800s, an American photographer named Oscar G. Rejlander created a composite photo of two people. The photo was called “The Two Ways of Life”, and it showed a young woman standing between two different paths in life. One path was dark and grim while the other path was light and happy.
Rejlander used multiple photos to create the final image. He took photos of the same woman standing in different positions and then combined them all together into one image. This was one of the earliest examples of deep fakery and it was met with a lot of controversy. People were surprised that it was even possible to create a fake photo like that, and they didn’t know how to respond to it.
Fast forward to the 21st century, with tools like Photoshop and AI programs, you can easily create fake photos and videos that look incredibly realistic.
Why you've heard it
Deep fakes and their applications, both good and bad seem to go hand in hand with the rise of social media. This is in part because it is easier to send information over the internet now more than ever because of social media. Part of the appeal of social media is your content appearing in front of as many people as possible and having them discuss it. The idea of a celebrity or famous political figure doing or saying something they never did can be used in both positive (nearly harmless) ways and in ways that can be truly devastating not only for the person being imitated but in the intentional spread of misinformation.
A "positive" example of deep fakes might be someone making a video showing a celebrity dancing to a song or singing along to the same song. It can also be used to show historical figures giving their famous speeches which can be a successful way for people to learn about those events without having to read about them in a book. These videos are mostly not meant to be harmful and cause little to no real damage to the person being impersonated.
However, you are also probably familiar with deep fakes due to their potential use in a variety of negative connotations. They can be used to spread disinformation which can be very hard to fact-check and find the original source after traveling throughout social media. Even if a piece of information may be easy to fact-check, many people will take deep fakes at face value (no pun intended), and will be hard pressed to think otherwise. This can be especially detrimental when it comes to politics and information about countries and their leaders as we have seen in numerous elections throughout the world.
Our Thoughts
As we have talked about, some forms of deep fakes have been around for a long time, but over the past few years, they have become more and more popular. This is in part due to the fact that deep fake technology has become increasingly more sophisticated over time, and it is now becoming easier and easier to create deep fake videos that can potentially go viral through social media. Because of this, they will most likely continue to advance and be used for good but mostly nefarious purposes.
They can be good for use in verifying the authenticity of videos or images. They could also be used to improve cybersecurity by helping to identify fake or fraudulent data. In addition, deep fakes could be used to create more realistic and lifelike videos for training, education and simulation purposes. Finally, deep fakes could be used to improve the accuracy of facial recognition software. Although, depending on who you ask, that can also be considered a bad thing.
While it is ideal to think of these beneficial uses, deep fakes will probably be used for a variety of nefarious purposes, such as spreading disinformation or compromising the security of individuals or organizations. For example, deep fakes could be used to create fake videos of political leaders making controversial statements, which could then be used to discredit them. In addition, deep fakes could also be used to launch cyber attacks by creating fake videos of employees logging into company networks from unauthorized devices, which could then be used to steal sensitive data.
Deep fakes also pose a serious threat to the security of individuals. For instance, deep fakes could also be used to create fake videos of people engaging in intimate or private activities, which could then be used to embarrass, humiliate, or blackmail them
Deep fakes also present a serious challenge for law enforcement and national security agencies. In this case, deep fakes could be used to create fake videos of witnesses giving false testimony, which could then be used to undermine the legal process. Alternatively, deep fakes could be used to create videos of people appearing to be engaged in criminal activity, which could then be used to blackmail them.
Cybersecurity experts are already aware of the dangers posed by deep fakes and are working to develop tools and strategies to combat deep fake threats. However, as deep fake technology continues to evolve, it will be increasingly difficult to safeguard against these threats. For the time being though, it is important for individuals and organizations to be aware of the dangers posed by deep fakes and to take steps to protect themselves against these threats which include:
First, be aware of the potential dangers and stay informed about the latest threats.
Second, use strong passwords and security measures to protect your online accounts.
Third, be skeptical of any information you see online, and verify the source before sharing it with others.
Finally, if you think you may have been targeted by a deep fake attack, report it to the authorities immediately.
Kotman Technology has been delivering comprehensive technology solutions to clients in California and Michigan for nearly two decades. We pride ourselves on being the last technology partner you'll ever need. Contact us today to experience the Kotman Difference.