What is Deepfake? Rashmika Mandanna Viral video: Just a few seconds of work for AI; What do people who hunt it do?

Table of Contents

What is Deepfake? Rashmika Mandanna Viral video: Just a few seconds of work for AI; What do people who hunt it do?

Rashmika Mandanna Viral video: The two pictures above are visible in the same frame. The first Instagram video is of Zara Patel’s real video, while the second video is of a fake video with Rashmika Mandanna’s face changed. The expressions seen in Rashmika’s fake video look absolutely real, so thousands of people on social media thought it was real.

From Bachchan to Rashmika Mandanna herself, everyone has been surprised. Rashmika wrote in a Twitter post, ‘To be honest, something like this is very scary not only for me, but for each one of us.’ I couldn’t even think of dealing with this if it had happened to me when I was in school or college.’

What is deepfake and how is it created?

The term deepfake was used for the first time in 2017. Later, videos of several famous people from deepfake IDs were posted on a social media aggregator called Reddit. It included several porn videos featuring Emma Watson, Gal Gadot and Scarlett Johansson.
Deepfake is the process of converting a person’s face, voice, and behavior into a real video, photo, or audio. This is so obvious that anyone can believe it. It also looks fake.
In this, the help of AI and machine learning is used. In this, video and audio are created with software and technology.
AI and cyber expert Puneet Pandey says that useful technologies and packages are now available. Now any person can use it. Current technology has also improved the voice. Sound cloning has become very dangerous in this.

How to make a deepfake video

First, such a photo or video should be identified in which fake face and voice can be found. Creating different types of faces is called encoder algorithm, while changing faces is called face swap.
The second program uses the decoder algorithm. Expression is applied when face swap occurs. To make the video appear real, audio is superimposed on each frame. One second of video is made up of 24 frames, or 24 pictures.
AI has made this even easier. Some public domain apps create a fake video or photo of the person by just inserting a photo. An example of this are many social video chat apps, in which you wear a live mask or crown to hide your face.
Experts say that anyone can get defeated by watching the video made with Perfect II technology. The victim’s image is tarnished unless the truth of the fake video is revealed. According to a study conducted in America, about 96% of AI-based deepfake videos are used in making porn.
Those making fake videos are two steps ahead of this. In 2018, when American researchers reported that fake videos did not move the eyelids, a few months later the eyelids also started moving in the fake videos
There is no certified tool yet that can differentiate deepfake videos, but many companies including Google, Amazon and Meta are continuously working on identifying deepfake videos. Still, the authenticity of the video can be seen by paying attention to a few things.
Those making fake videos are two steps ahead of this. In 2018, when American researchers reported that fake videos did not move the eyelids, a few months later the eyelids also started moving in the fake videos.
There is no certified tool yet that can differentiate deepfake videos, but many companies including Google, Amazon and Meta are continuously working on identifying deepfake videos. Still, the authenticity of the video can be seen by paying attention to a few things.

There is no separate law to deal with this in India.

The European Union has implemented a “Code of Practice on Disinformation” under its AI law to prevent deepfakes. Following the signing, several tech companies including Google, Meta and Twitter will have to take strict action to prevent fake accounts and deepfakes on their platforms.

There is six months time to implement. For breaking the law, the company will have to pay a fine of 6% of its annual world revenue.

US Disaster Task Force Act is in force. It’s not too harsh yet. After this is the annual analysis of deepfake technology. The US states California and Texas have also passed separate laws.

There is no law for this in India. Only the IT Act hears such cases.

Anuj Kumar Aggarwal, an expert in cyber matters and Supreme Court advocate, says that there is no separate law for this in India. Anyone spreading such videos or information can be sentenced to three to ten years of imprisonment under the IT Act.
An amendment in 2008 has given exemption to social media companies on the condition that they cooperate with the investigating agencies in such cases, otherwise no legal action will be taken against them.

Latest Job Update

IMPORTANT Notice-

“Daily updates at your fingertips. Don’t miss out on what’s happening in the world. Get your daily news, jobs, entertainment, sports, and business updates.”Seachgyan is the trusted source of these updates.

Searchgyan

Check it now.

Thanx For reading our article .Team Searchgyan wishes you a bright future.

Connect with us on-

Facebook

Twitter