DEEPFAKES

 Deepfakes are media that are synthetic and have been altered digitally to replace a person’s likeness with that of another very convincingly. Why should you and I care about deepfakes? In this article from WIRED UK (read it here- The problem with deepfakes - if you are interested), the author, Rachel Botsman, talks about the biggest problem with deepfakes is that people don't care what's real. She uses her son as an example. She talks about how he came home from school one day and asked her what a deepfake was. He was eight years old when this happened. Since this was in 2019 there were a lot of Donald Trump videos going around with Trump saying things he didn’t actually say that everyone that was on the internet saw. Rachel had to explain to her son that just because you see a video of someone saying something these days, does not mean they actually said it. 

That type of logic is very difficult for an eight year old to understand and frankly, they don’t care. The sad part is that the average human being doesn’t care. Many people if they see a video of someone saying something will always hold on to the belief that that person actually said it, even if it is proven to them that that person did in fact not say what was said in the video. While the idea of making fake content is not new, deepfakes use very effective and powerful techniques from artificial intelligence and machine learning to manipulate or change visual and audio content so that it can more easily deceive the average human brain.

Rachel goes on in her article to talk about this idea that fake content is not new. She uses Photoshop (National Geographic and other magazines change images so that they fit the cover better than the actual picture that was taken) and consumer-generated imagery (Tom Hanks meeting meeting President Kennedy in Forrest Gump, as we all know that didn’t actually happen in real life) as examples of how it has always been hard to tell what is real and what is not real online. 

This wonderful article by Rachel Botsman I keep referencing I found when I was doing my research for my presentation for this class for deepfakes. It taught me so many things about deepfakes that I really did not know about before. However, I thought the most interesting thing was how Mrs. Botsman talked about the problem with deepfakes is really not all the technology stuff. It is more about the human aspect of the whole thing. As people, our brains are created to see what we want to see or believe. This idea is essentially the desirability bias. We believe something and then go look for how to prove that belief to others. It is exactly like writing a paper taking a side in an argument and then looking for sources that you can talk about in your paper that back up your claim.

Deepfake technology has been developed by researchers at certain academic institutions beginning in the late 1990s. The Video Rewrite Program in 1997 was the first program to alter existing video footage to create footage of someone saying or doing something they didn’t do previously. Deep fakes have later more recently been developed by amateurs in online communities such as Reddit. Recently the methods of deep fakes have been adopted by industry and we are seeing a lot more of artificial intelligence in our everyday lives.