Why has the Government Issued a Directive on Deepfakes?

Relevance: GS3 – Science & Technology, Artificial Technology, Deepfake Technology, Impact of Deepfake Technology, India's stand on dealing with deepfakes, Ethical concerns associated with deepfakes, Information Technology, Cybersecurity

Primary Source: The Hindu

Click here for Daily Current Affairs

Why in the news?

• Recently, the Government of India has mandated "social media intermediaries" to remove deepfakes within 24 hours following a complaint, aligning with the IT Rules 2021.
• This action follows after deepfake videos of actors Rashmika Mandanna and Katrina Kaif emerged online.

What are Deepfakes?

• Deepfakes are highly realistic and often convincing digital forgeries of videos, images, or audio recordings.
• They are generated using artificial intelligence (AI) and machine learning techniques, particularly deep learning, from which the term "deepfake" is derived.
• Deepfakes are made using algorithms called generative adversarial networks (GANs), among other techniques.
• GAN consists of two parts: the generator, which creates images or videos, and the discriminator, which evaluates their authenticity. They work in tandem to improve the quality of the forgeries.
• With deepfake technology, creators can make people appear to say or do things they never did.
• This includes facial expressions, lip movements, voice, and even body language if full-body deepfakes are made.
• A 230% increase in deepfake usage by cybercriminals is noted. Experts predict deepfakes might soon eclipse phishing.

What is the Technology Behind Deepfakes?

• The technology behind deepfakes primarily involves a form of artificial intelligence known as deep learning.
• This is further elaborated using a specific machine learning technique called a Generative Adversarial Network (GAN).
• Deepfake technology uses GANs, which consist of two parts - a generator and a discriminator. These two parts work in tandem to create and refine deepfakes.
• The generator creates images or videos that are intended to be indistinguishable from authentic ones.
• The discriminator assesses these creations for authenticity, identifying signs of forgery.
• The AI-driven software learns the subjects' movements and facial expressions from source material (such as photos or videos of the person to be deepfaked).
• It then replicates these features in another video or image, creating the deepfake.
• To make the deepfake as realistic as possible, a large database of source images is typically used, especially for public figures, celebrities, and politicians, due to the abundance of available images.
• The fake video or image is continuously refined until the discriminator part of the GAN can no longer detect any signs of forgery.

Laws Against Deepfakes in India

India does not have any specific laws that exclusively address deepfakes. However, various provisions of existing laws can be applied to tackle issues arising from deepfakes.

• Section 66E of the IT Act of 2000: This section applies in cases of deepfake offenses that entail recording, publishing, or sending a person's photographs in mass media, infringing on their privacy.
  • Offenders face up to three years in prison or a fine of up to two lakh rupees.
• Section 66D of the IT Act of 2000: It allows for the prosecution of those who utilize communication devices or computer resources with the criminal goal of cheating or impersonating someone.
  • It is punishable by up to three years in prison and/or a fine of up to one lakh rupees.
• Copyright Protection: The Indian Copyright Act of 1957 protects works such as films, music, and other creative content.
  • Copyright holders can sue persons who construct deepfakes utilizing copyrighted works without permission.
  • Penalties for copyright infringement are set out in Section 51 of the Copyright Act.
• Government Advisory: On January 9, 2023, the Ministry of Information and Broadcasting issued a warning to media outlets to be cautious when airing content that could be modified or tampered with.
  • The Ministry also suggested that manipulated content be labeled as "manipulated" or "modified" to alert viewers that the content has been altered.

Global Initiatives to Combat Deepfakes

• European Union: To combat the spread of disinformation by deepfakes, the European Union amended its Code of Practice on Disinformation in 2022.
• United States: The bipartisan Deepfake Task Force Act was introduced in the United States to aid the Department of Homeland Security (DHS) in combating deepfake technology.
• China: It will implement comprehensive deep synthesis regulations beginning in January 2023. The law, which aims to reduce deception, mandates unambiguous labeling and traceability of deep synthesis content. Individual consent is required, as is obedience to laws and public morals. Service providers must put in place review systems and work with authorities.

Positive and Negative Aspects of Deepfakes

Positives

• Assistance for ALS Patients: Deepfakes have been used positively in cases like the collaboration between the ALS Association and a company to help people with ALS digitally recreate their voices.

Negatives

• Increase in Criminal Usage: There has been a 230% rise in the use of deepfakes by cybercriminals and scammers.
• The technology, which is becoming increasingly accessible, allows for the creation of convincing fake content that can be used for malicious purposes.
• Potential Replacement of Phishing: Cybersecurity experts predict that deepfake technology could soon replace phishing as a primary method of online fraud and deception.
• Impersonation and Misinformation: Deepfakes pose a significant threat in terms of impersonation, potentially leading to political instability, sabotaging elections, unsettling diplomatic relations, and spreading misinformation.
• Difficulty in Detection: The technology, particularly through the use of Generative Adversarial Networks (GANs), has evolved to a point where it becomes increasingly difficult for software to detect these fakes.
• Misuse Against Individuals and Organizations: There is a risk of deepfakes being used to humiliate or blackmail individuals, as well as attacking organizations by presenting false evidence.

Conclusion

Deepfakes represent a significant technological advancement with potential for both harm and good. The response by the Indian government, alongside international efforts, highlights the need for vigilance and regulation in the age of synthetic media.

(*Click this link to read prelims specific weekly current affairs articles)

FAQs

MCQs

Question: With the present state of development, Artificial Intelligence can effectively do which of the following? (UPSC 2020)

1. Bring down electricity consumption in industrial units
2. Create meaningful short stories and songs
3. Disease diagnosis
4. Text-to-speech conversion
5. Wireless transmission of electrical energy

Select the correct answer using the code given below:

(a) 1, 2, 3 and 5 only

(b) 1, 3, and 4 only

(c) 2, 4, and 5 only

(d) 1, 2, 3, 4 and 5

Question: In India, it is legally mandatory for which of the following to report on cyber security incidents? (UPSC 2017)

1. Service providers
2. Data centers
3. Body corporate

Select the correct answer using the codes given below:

(a) 1 only

(b) 1 and 2 only

(c) 3 only

(d) 1, 2 and 3