Modern technology allows the storage and mapping of your face to photos and videos. Here's what you need to know about this new potential threat.
What is Deepfake Technology?
Deepfake Technology or “deepfakes” are manipulated photographs or videos, which rely on artificial intelligence (AI)-enabled technologies to splice the likeness of someone onto the face of another. It can manipulate images or videos of someone to make it seem as if they are doing or saying something that has never happened. The part of the name “deep” comes the term “deep learning,” which is an AI feature. By learning about your face and facial expressions, and with as little as one minute of conversation, these deep-fake programs can recreate your likeness and provide visual representations of you – ones that are fake.
The technology behind this software has expanded rapidly since 2017, as programs have become easier to use, better quality and much cheaper than they were in their inception. Now, people all over the world are able to use this new type of technology – sometimes, people will use it willingly through third party applications created to collect and store your information.
New applications such as FaceApp, created by Russian-based company, Wireless Lab, became a trend due to its ability to age the look of your face. Soon after going viral, the app sparked privacy concerns due to the fact that by using the app, users are allowing Wireless Lab to store and own their likeness as outlined in the app’s terms and conditions. This means that a user’s images could be used in advertisements, or, more dangerously, for fake content. These deep fake programs also have the ability to use one source image before manipulating facial expressions and more, as demonstrated with the famous portraits below.
Why is Deepfake Technology Dangerous?
Best case scenario, it can be a fun joke between yourself and friends. Worst case scenario, someone has projected your likeness onto another or produced content of you doing or saying things you have not done. The potential to create fake news, violate someone’s privacy or obtain secure information is much more at risk now than ever before.
To see just how real it is, watch this video created using deepfake technology:
It looks and sounds like former US President Barack Obama. However, it is actually the comedian and filmmaker, Jordan Peele, providing the template to produce a speech which Obama was not responsible for. Other celebrities, such as Ariana Grande and Jennifer Lawrence, have found themselves as the target of deepfake technology for the use of X-rated content, with perpetrators transferring their likeness into explicit videos. Twitter and the pornographic distributor, Pornhub, have both banned such technology to be used in their posts. However, widespread regulation has been minimal so far.
Facebook has openly stated that they will continue to allow deepfake content and will not remove videos or images which employ AI technology. In response, a recent Instagram post discussed the implications of such a policy, namely because of how much information is available through Facebook.
It’s not just celebrities being subjected to the usage of this technology either, as ex-partners have learned to use deepfake software to create “revenge porn” targeting their previous partner, often using these images and videos to torment or post publicly.
In other cases, it has been used to politically influence people as well. Nancy Pelosi, a United States Speaker of the House, had a speech of hers manipulated to make her sound inebriated, of which President Donald Trump used on social media, which was later proven to be fake.
One of the major concerns regarding deepfake videos is the potential to be used during live video presentations, which appears to be the next step in its evolution. With that advancement in deepfake technology, it will allow users to impersonate and manipulate video of others during live presentations.
In an age where real news and information is often overshadowed by distracting and fake news, it’s important to be cautious and critical of what you are seeing on the internet.
What can you do to protect yourself from deepfake technology?
Do not subscribe to apps that store your facial likeness
If you are the victim of revenge porn, ask the host to remove it
Revenge porn is illegal in Canada (See Bill C-13) and some states in America. If the host will not remove the post, find out the laws in your state or find an anti-cyber bullying resource who can put you in touch with resources.
Only trust videos and images from credible resources
Use your better judgement. Do not trust images or videos from sources you don’t recognize.
Engage with content before you believe it as absolute truth.
It is cliché, but don’t believe everything you read on the internet! Make sure to double-check sources and information. Research is your friend!
It is a complicated time in the modern age, and technology continues to surpass our expectations.