Blog Post #9 (Key Post)

     Through further research for my EOTO presentation, I have begun to realize that deepfakes can be a fun and useful tool. It's a great advancement in technology that stores face data and can be used for good purpose. However, that is actually the problem with deepfakes, is the fact that it's not being used for good purpose. Because of this, deepfakes can be very threatening to almost anyone. 

    The term deepfake also means "Deep Learning Fake" referring to a specific learning algorithm that utilizes artificial images and sounds to create fake pictures or videos with someone else's likeness or voice. It's pretty much a super advanced photo editing software. An interesting fact is it was originally made by a reddit user manipulating Google's open source technology to make inappropriate content. With the use of a Generated Adversarial Network or GAN, it uses a specific set algorithm to make, combine, and cluster fake images and try to impose them on ones face matching it up, superimposing it on ones body and attaching a voice to it to create videos or pictures of one's likeness.

    Unfortunately, this seemingly astonishing piece of technology has been misused from the very beginning. The best thing it's known for today is being a primary tool for media manipulation. Deepfakes were used for different media such as Online Scams, Pornography, Identity theft, Financial fraud, and most importantly,  Election manipulation, as well as just spreading disinformation in general. The only thing it was used in a positive way was in animated movies

    All of these reasons that I've listed above could more or less explain how deepfakess are a real threat to our society, but the real question would be WHY deepfakes are a real threat. The main reason is because of who deepfakes can affect, which is unfortunately everyone, young and old. This goes especially for those who are connected to the internet and social media. Because of deepfakes and it's heavy media influence, it can c ause a lot of controversy for unexpecting people, going on very touchy subjects like politics, using someone's face or voice to spread disinformation. Also, audio deepfakes can play a part in financial scaams or taking your identity and can threaten internet security, and most people aren't really aware of whether or not they're listening to a deepfake.

    I feel that because of these deepfakes that we need to further exercise cautions as we take information from certain sources. Besides finding out whether or not the sources we use are credible, there are some way we can tell if a piece of media is a deepfake. The key is to try to find anything unnatural looking, such as their eye movement, skin color, positioning of their head and body, terrible lip syncing, and peculiar voices, among other things. Hopefully i n due time, we might come up with a technology y to counter these deepfakes, but it's better to be safe and make sure the sources which you get your information from are trustworthy.