Actress Rashmika Mandana is currently in the news and her deepfake video is going viral on social media.

A deepfake viral video of Rashmika, who made her mark with a successful film like ‘Pushpa’, has now created a huge controversy.

An attempt has been made to show that the woman seen in this video is Rashmika Mandana through deepfake video.

Rashmika has expressed her displeasure over this and has appealed to find a solution at the earliest so that no one else suffers like her.

Rashmika wrote on social media platform X, “Honestly anything like this is very scary not just for me but for all of us.”

She further writes that the way technology is misused today, it can cause great harm not only to her but also to many others.

Rashmika writes, “Today as a woman and an actress, I am fortunate. Because my family, friends and well-wishers are my protectors and support system, and I am thankful to them.”

“But if something like this had happened when I was in school or college, I can’t imagine what the situation would have been or how I would have dealt with it.”

How do you know if a video is deepfake? What did Amitabh Bachchan say?

Citing this video, actor Amitabh Bachchan has said that legal action should be taken in this matter.

Meanwhile, Union (State In-Charge) Minister Rajeev Chandrasekhar said social media platforms should ensure that wrong information is not shared on their platforms.

This viral video is a deepfake, according to a fact checker.

Associated with fact-checking website Alt News, Abhishek said on X (Twitter) that, “The woman seen in this video is not Rashmika Mandana but this video was made using deepfake technology.”

What is Deepfake?

Deepfake is a technology that uses AI technology to alter videos, photos and audio.

With the help of this technology, the face of a person can be superimposed on a person’s photo or video.

Simply put, artificial intelligence or AI technology is used to create fake videos that look real but are actually fake. That is why it has been given the name Deepfake.

According to a report, the term became popular in 2017 after a Reddit user used the technique to change faces in pornographic videos. Reddit later banned ‘deepfake porn’.

How does this technology work?

Deepfake is a very complex technology. For this one should have proficiency in machine learning i.e. computer.

Deepfake content is created using two competing algorithms.

One is called decoder, and the other is called encoder.

In this, fake digital content is created, and the decoder asks if the video is fake. It is made to find out.

Each time the decoder correctly identifies whether the content is real or fake, it sends this information to the encoder. so that the encoder can correct errors in deepfake.

Combining both processes results in a generative adversarial network called a GAN.

Where are deepfakes used?

According to reports, this technology started with creating obscene content.

This technique is used a lot in pornography. Obscene content is posted on porn sites using the faces of actors and actresses.

According to a Deep Trace report, 96 percent of deepfake videos found online in 2019 contained pornographic content.

Apart from this, this technology is also used for entertainment. These deepfake videos were used to show that ‘nothing really happened’ so that viewers would believe it.

Many YouTube channels post deepfake videos of scenes from various movies.

For example, a deepfake video of a famous scene from the movie ‘The Shining’ is available on YouTube channel Ctrl Shift face.

Over the past few years, this technology has also been used to relive the past (nostalgia). For example, faces are animated in photographs of dead relatives.

People used this feature to bring their ancestors and historical figures to life through technology.

​ Deepfakes are now being used in politics as well. In elections, political parties criticize each other through deepfake technology. Some deepfake videos surfaced during the Ukraine-Russia war.

 How to identify deepfake content?

To identify deepfake content, it is important to pay attention to certain things.

The condition of the face is first seen in them. Deepfake technology often hits the face and eyes. It also observes the movement of the eyelids.

If you feel that the eyes and nose are moving somewhere else or the eyelids do not move for a long time, then understand that this is deepfake content.

The color scheme is looked at to identify whether a particular photo or video has been tampered with and identifies deepfake content.