Deepfakes go Mainstream

“Not every video on the Internet is real and the fake ones are multiplying. That’s thank to the spread of Deepfakes. Deepfakes are videos that have been altered using machine learning, a form of artificial intelligence, to show someone saying or doing something that they did not in fact do or say. The results can be great fun. Take for example these hilarious examples of clips that show Nicholas Cage starring in movies that he was never in, but deepfakes can also be a tool for harassment and a way to spread political misinformation.”

Deepfakes in not just Face swapping! Check the following videos!

As Wikipedia mentions “Deepfake (a portmanteau of “deep learning” and “fake”[1]) is a technique for human image synthesis based on artificial intelligence. It is used to combine and superimpose existing images and videos onto source images or videos using a machine learning technique known as generative adversarial network.[2] Because of these capabilities, deepfakes have been used to create fake celebrity pornographic videos or revenge porn.[3] Deepfakes can also be used to create fake news and malicious hoaxes and financial fraud.[4][5][6]

[1] Brandon, John (16 February 2018). “Terrifying high-tech porn: Creepy ‘deepfake’ videos are on the rise“. Fox News. Retrieved 20 February 2018.
[2] Schwartz, Oscar (12 November 2018). “You thought fake news was bad? Deep fakes are where truth goes to die“. The Guardian. Retrieved 14 November 2018.
[3] “What Are Deepfakes & Why the Future of Porn is Terrifying“. Highsnobiety. 20 February 2018. Retrieved 20 February 2018.
[4] “Experts fear face swapping tech could start an international showdown“. The Outline. Retrieved 28 February 2018.
[5] Roose, Kevin (4 March 2018). “Here Come the Fake Videos, Too“. The New York Times. ISSN 0362-4331. Retrieved 24 March 2018.
[6] “Adversarial Learning of Deepfakes in Accounting” (PDF). Arxiv.org. Retrieved 9 October 2019.


So all these facts really got my attention and as an Information Security enthousiast I was really concerned about the unethical usage of this technology. Deep fakes could be used for Social Engineering engagements, Scamming, Fraud, Political propaganda, Spreading fake news even Sextortion.

So is it that difficult to create your own deep fake video? I have little to zero experience and skills related to video/audio editing so I was pretty discouraged. Although after a short
research I found this GitHub repository called DeepFaceLab (https://github.com/iperov/DeepFaceLab) “DeepFaceLab is a tool that utilizes machine learning to replace faces in videos.” It is opensource, uses NvidiaCUDA, OpenCL, Keras, TensorFlow, PlaidML and has a pretty detailed documentation.

Unfortunately running this requires a significant hardware resources especially a strong GPU. Not the best description of my poor old Dell laptop.

So I found this website called Deepfakesweb (https://deepfakesweb.com) that promises to do the same thing on the Cloud for you and you just pay a small amount for the approximately 4-5 hours that will be needed for the training of your model and the video rendering.

Using the service is extremely easy. Even a non-tech user can definitely use it. It is just 3 steps!

(Although I am not promoting this website I am really suggesting you give it a try)

For my experiment I downloaded two high definition videos from YouTube. The first is the prime minister of Greece and the second on the Minister of Labour. Two really well known Greek political figures and that is why I choose them.

Original 1 (Prime Minister)

Original 2 (Minister of Labour)

Final Product

 

 

Keep in mind that spreading Deepfakes videos is prohibited by law in some countries and regions so be really careful with that!

I think the final result is pretty impressive. Image if a one-man-squad can do this pretty decent video in 4 hours with a few bucks over and old laptop what a well funded political organization or group of experts in cooperation with good actors can do if it is needed!

Just some food! 😉

Leave a Reply

Your email address will not be published. Required fields are marked *