Just how Dangerous are Deepfakes?
Are deepfakes as dangerous as they sound?
Let’s face it Deepfakes have been around for a while. Governments have been propounding propaganda for years faking all kinds of pictures and even video using old-school techniques.
The rise of image synthesis using specific facial features has made the creation of all kinds of fake pictures, videos and even audio!
Check out this survey concerning deceptions and deepfakes on the internet.
This has become a worrisome trend and it will affect how people see things; especially in an age where social media rules and because we exist in an age of social media where people can post whatever they want, the effects cause problems that a lot of the time can’t be controlled.
Today, all kinds of news become viral in seconds.
From what we think is the latest technological news to the latest political news.
New trends such as the latest cryptocurrency news to the latest religious news.
Many times, such news items are fake.
With deepfaking, it is going to get much worse!
How Exactly Does Deepfakes Work?
The human face is made up of pixels when it becomes a digital image.
These pixels can be learned and copied using methods that artificially intelligent software can copy in minutes.
With what is called an artificial neural network that functions very much like the human brain processing all kinds of data everything now falls into place.
This process referred to as human image synthesis has been around for years but most of us didn’t bother about it.
It was a Reddit user called deepfake that created a software program around the process.
He developed an application called a fake app that can copy the process in literally minutes with the right hardware.
What About Audio?
It is possible to also create audio that is fake and that can’t be authenticated on time due to the daily pressures of life.
Descript which is an artificial intelligence startup has acquired an application (Lyrebird) that does this quite easily.
Putting the two together can create some potentially dangerous situations.
And it is already happening.
Bizzare Possibilities and Scenarios
We live in politically charged times. Many countries and their citizens ( the United States inclusive) rely on social media for being able to motivate people and spread their message.
Political revolutions these days aren’t started on the ground.
They are started online.
With extremely sensitive content that can tell the story right, anything can become possible.
This brings to light the question which everyone is afraid to ask: what if a series of politically sensitive videos are released at once and spread at a rapid pace en masse?
These are the kinds of scenarios that cause revolutions overnight!
On the other hand, a popular politician in any country can be brought down with a series of videos and audios that may depict him or her in a certain light.
At the end of the day, such an individual may be proven innocent but the damage would have already been done.
This can be applied in so many different scenarios: business, religion, social affairs and so on.
Financial markets can be manipulated in second without actually influencing them on the inside.
If the people (especially retail investors) can be manipulated then such financial markets can also be played as well.
By the time such “fake news” has been countered, the profits have been made and the masterminds behind such a scenario will disappear only to reappear again to do the same.
These and many more are the kinds of possibilities that present a stark future in an age where propaganda is more important than the truth.
This brings us to an important question.
Can Deepfaking be countered?
In theory, it will almost be impossible to capture most of the scenarios which have been described.
In practice, however, it is very possible to curb such scenarios to a significant degree.
Community-oriented solutions for fact-checking for starters will serve as a check on incoming material.
State-sponsored fact-checking and How Big Data Can be used to Improve Human Existence solutions also factor in this model.
Corporate technology giants that deal with content daily which include search engine giants, social media giants, forums and other platforms to name a few can help counter this problem.
News organizations also need to put together fact-checking units to verify every single piece of content before publishing or broadcast.
Social media giants can counter this problem by creating deep learning and highly intuitive neural engines which can detect in real-time potential fake content.
Rules must be applied across the board.
Effective laws, rules, and regulations can also be put in place to make sure that there are punishments and corrections meted out to offenders.
Proper enforcement of such laws, rules, and regulations by governments, communities (both online and offline), corporate bodies, media organizations and everyone else must adhere to the rules.
This way, fake news, fake events and fraud of such an advanced nature can be countered.
Putting a Lid on the Problem
While there should be no limits on creativity, it is important to ensure that everyone is playing by rules of etiquette and ethics.
We are entering a decade where criminals will become more sophisticated and daring.
What many may think is truth will be lies.
Many will fall for it.
If systems are put in place to ensure the sanity of content, then this problem shall be contained.
What do you think? Are deepfakes dangerous?
Please let us know in the comments below!