What are Deepfakes?

About Deepfakes

• Deepfakes are a compilation of artificial images and audio put together with machine-learning algorithms to spread misinformation and replace a real person’s appearance, voice, or both with similar artificial likenesses or voices.
• The term "deepfake" combines the deep learning concept with something fake.
• It can create people who do not exist and it can fake real people saying and doing things they did not say or do.
• Background: The origin of the word “deepfake” can be traced back to 2017, when a Reddit user with the username “deepfakes”, posted explicit videos of celebrities. 
• Working:
  • They are created by machine learning models, which use neural networks to manipulate images and videos. 
  • To make a deepfake video of someone, a creator would first train a neural network on many hours of real video footage of the person to give it a realistic “understanding" of what he or she looks like from many angles and under different lighting.
  • Then they'd combine the trained network with computer-graphics techniques to superimpose a copy of the person onto a different actor.
• Deepfake technology is now being used for nefarious purposes like scams and hoaxes, celebrity pornography, election manipulation, social engineering, automated disinformation attacks, identity theft, and financial fraud.
• Deep fakes differ from other forms of false information by being very difficult to identify as false.

Q1) What is machine learning?

Machine Learning, often abbreviated as ML, is a subset of artificial intelligence (AI) that focuses on the development of computer algorithms that improve automatically through experience and by the use of data. In simpler terms, machine learning enables computers to learn from data and make decisions or predictions without being explicitly programmed to do so.

Source: New regulations to tackle deepfakes soon: IT Minister Vaishnaw