Cyber Security – CERTs, Policy, etc

What are Deep Fakes?

Note4Students

From UPSC perspective, the following things are important :

Prelims level : Deep Fake

Mains level : Cyber bullying and other threats posed by AI

Cybercrime officials in India have been tracking certain apps and websites that produce vulgar photographs of innocent persons using Artificial Intelligence (AI) algorithms. These images are then used to blackmail victims, seek revenge or commit fraud on social networking and dating sites.

The most notorious misuse of AI is knocking the door. The Deepfake is an application of Deep Learning (an axiom of AI and Machine Learning). UPSC may ask a mains question about the challenges posed by AI-based technology.

What is Deep Fake?

  • Cybercriminals use AI software — now easily available on apps and websites — to superimpose a digital composite (assembling multiple media files to make a final one) on to an existing video, photo or audio.
  • They are computer-generated images and videos.
  • Using AI algorithms a person’s words, head movements and expressions are transferred onto another person in a seamless fashion.
  • That makes it difficult to tell that it is a deepfake unless one closely observes the media file.

Threats posed

  • Because of how realistic deepfake images, audio and videos can be, the technology is vulnerable for use by cybercriminals who could spread misinformation to intimidate or blackmail people.
  • With real-time face tracking it is becoming easier to fabricate believable videos of people doing and saying things they never did.
  • There are rising cases of “revenge porn” i.e. creation of sexually explicit videos or images that are posted on the Internet without the consent of the subject as a way to harass them.

What are the catfish accounts?

  • Catfishing refers to the practice of setting up fictitious online profiles most often for the purpose of luring another into a fraudulent romantic relationship.
  • A “catfish” account is set up a fake social media profile with the goal of duping that person into falling for the false persona.

What can we do to protect yourself?

  • A basic check of their social media profiles, comments on the images and whether similar profiles exist could help determine if the person is genuine.
  • While it is not easy to keep track of who downloads or misuses the user images, the best way to protect is to ensure that we are using privacy settings on social media profiles.
  • If one feels his/her image has been used without prior permission, they could use freely available reverse image search tools to find images that are similar to yours.
  • One can also be mindful of who he/she is conversing with on the web.
Subscribe
Notify of
0 Comments
Inline Feedbacks
View all comments