Written by Mahanika Himalayay Upadhyay, third year student pursuing BA LLB at Christ (Deemed to be) University, Delhi NCR

Deepfakes refers to the fraudulence which is created by deep learning of AI and its subsets. It can create entirely new content by manipulating the content which already exists. They can be in the form of  images and videos which appear incredibly realistic and are made using face-swapping technologies. AI and deepfakes specifically have shown remarkable growth as for the first time it appeared in the end of 2017. Deepfakes are of  many types, for example – face swapping, re-enactment, lip syncing, motion transfer, image generation, audio generation and even text generation. It has abounding advantages and disadvantages. This blog breaks down the ways in which deepfakes endanger personal consent, damage reputation, and weaken democratic principles. It also considers whether current legal frameworks are capable of controlling these type of abuses. Moreover, the blog makes a case for the necessity of creating specific and fair regulations to address the issue.

Notably, such low technical skills and equipment are required for the creation of deepfakes that anyone can easily produce and distribute them online. Deepfake offers a spectrum of benefits across various fields, which shows a wide variety of  possibilities which have a  positive impact on the society. It is advantageous in the entertainment, medical and education industry.

In the entertainment industry, deepfake has brought a revolution in filmmaking as it has enabled computer-generated imagery and replication of digital character which appear extremely realistic.

In the case of medical industry, with the help of deepfake, lifelike scenarios and patient interaction can be easily replicated without any risk to real patients. It offers valuable experience for the ones who practice medicine.

 Academically, deepfakes have the possibility to create engaging learning exposure. Likewise deepfake has an abundance of advantages.

The influx of deepfake technology has introduced perpetual risks. These risks extend to numerous domains which pose serious challenges to society. The fundamental concern is the destruction of information purity. It is very burdensome and challenging to differentiate between what is original and what is fake. The high possibility of identity theft and cybercrime has intensified security concerns. The technology may also sabotage the reliability of evidence in courts. The validity of digital content is also affected in court proceedings.

Another dangerous threat posed by the misuse of deepfake is to the democratic system, which was till now free from electoral manipulation. Deepfake’s use in election campaigns has now become common in many countries. For example, in Indonesia the voters could see and hear late President Suharto who encouraged them to vote by using deepfakes. Another example is from Slovakia, in the parliamentary elections last year, during the time of campaign, two days before the vote, a voice recording went viral on Facebook in which Progressive Slovakia Party’s leader, Michal Šimečka allegedly  discussed about election fraud to a journalist. The usage of deepfakes was also seen during the last elections in India when on the eve of the vote, a recording in which Manoj Tiwari, the President of Delhi state from BJP, allegedly criticised his political opponents in a released recording. Also, in 2023 a video circulated on Instagram of actress Rashmika Mandanna, where a video of British-Indian woman Zara Patel was face-swapped with her’s. Information Technology Minister of India regarded these type of videos as a more “damaging form of misinformation”. These cases are proof that the misuse of deepfake is an extremely serious and dangerous issue not only nationally but globally.

There are no definite laws addressing the issue of deepfakes in India. Due to the diversity of the usage of deepfakes, the wrongs and harms caused by them can be brought under current legal structures. A deepfake which is defamatory may invoke defamation laws like Section 358 of Bharatiya Nyaya Sanhita 2023.[1] A deepfaker can be penalised under Section 271(2)(a) and/or Section 75 of Bharatiya Nyaya Sanhita, or the Indecent Representation of Women (Prohibition) Act 1986 for creating a non-consensual pornographic deepfake.[2] Since definite deepfake legislation/regulation is lacking in India, courts and scholars seem to be deviated towards jurisprudence of publicity law to address the harms caused by deepfakes. Publicity protection is addressed in Indian courts under Article 19(1)(a) and 21 of the Indian Constitution (freedom of expression and right to life)[3] and IP laws. Other laws which indirectly deal with punishment of deepfakes include Section 111, 319, 358 of the Bharatiya Nyaya Sanhita, 2023; Section 66C, 66D, 67 of the Information Technology Act, 2000; Rule 3(1)(b)(viii); Rule 3(2)(b) of Information Technology Rules, 2021 and Section 6 of Digital Personal Data Protection Act, 2023.[4] There are regulatory advisories also regarding this, these are-  November 2023 advisory and ECI directives of May 2024. Through the case of K.S. Puttaswamy v. Union of India,[5] the Supreme Court held that right to privacy is the essence of right to life and personal liberty as in the Article 21 of the Constitution of India. Judicial take on deepfakes include Gaurav Bhatia v. Naveen Kumar & Ors.,[6] National Stock Exchange v. Meta Platforms and Ors.,[7] Karan Johar v. Ashok Kumar, John Doe and Ors.[8] and Suniel Shetty v. John Doe[9].

Since there are no specific laws for combating deepfakes, the current laws pose many challenges. These are:

1. Lack of distinction – punishment of harm caused by a deepfake has to be settled only by extensive judicial construction. It is because there is no proper law which addresses the misuse of deepfake. Vicious use of deepfake has far-reaching consequences as a person’s reputation may be tarnished and even there are high chances of national security threat.
2. Issues related to enforcement – while creating deepfake is very easy but its detection and proof collection using digital evidence requires higher technical knowledge. Therefore, due to lack of necessary tools and low cyber literacy it is very difficult to curb deepfake-enabled cybercrime.
3. Jurisdictional challenges – many times deepfake related cybercrimes are transnational which require global cooperation in order to prosecute the offenders. International politics affects the execution of this. Also, in India Bharatiya Nyaya Sanhita states that every person who commits crime within the territory of India shall be held liable under this code. But it is nowhere stated that what constitutes a crime within India in relation to deepfake cybercrimes. Hence there is extreme difficulty in enforcing and prosecuting in these situations.
4. Privacy and Consent Issues – In Facebook India Online Services Pvt. Ltd. v. CCI,[10] privacy policy of WhatsApp certified that data of its users is shared with Facebook, and its non-agreement would result in discontinuation of its services. The Delhi High Court stated that it was irrational and unjust to the users of WhatsApp.

Major nations have already taken Regulatory approaches towards the misuse of deepfakes. Some of them are given below-

1. United States of America: In the USA, many proposals regarding deepfake has been introduced like Deepfakes Accountability Act and the Protect Elections from Deceptive AI Act but there is no specific federal legislation addressing deepfakes. The proposals above have not been evolved into a law yet. But at the state level, legislation which addresses deepfake have been passed in 23 states as of 2024, out of which 17 states have enacted bills or amendments.
2. China: In January 2023, Administration of Deep Synthesis of Internet Information Services was adopted to regulate deepfake content. One of the fundamental requirements of this law is that it is mandatory for all AI generated content to be labelled. Also, without the consent of the user, deepfake production of them is illegal. Tech firms in China like Alibaba and Tencent are obligated to align with the government’s requirements.
3. The European Union: An act designed to cultivate trustable AI along with addressing future risks which are associated with AI generated content has been introduced by the EU called Artificial Intelligence Act. This act makes it mandatory for the creators or users of deepfake to be transparent and disclose that the origin of the content is artificial, also to provide information on which technique was used in its creation. It also differentiates the risk level of AI systems into four categories: unacceptable, high, limited and minimal.

Similarly, India must also introduce a regulation or an act in order to protect the innocent from the malicious use of deepfake. This would ensure that the fundamental right to life and personal liberty as in Article 21 is not violated due to the misuse of new technologies like AI and specifically deepfake.

An act specifically addressing the issue of deepfake, preventing its malicious use and also regulating it should be introduced by the government. Content produced by AI or any alteration done without permission should be regulated and provisions which provides punishment for such misuse should be initiated by the governments. A specific branch in the Indian Computer Emergency Response Team (CERT-In) should be created so that the online content can be reviewed if it is original or deepfake.

In order to combat deepfake, online platforms can also take these steps:

1. Awareness and education of the users: The platforms can educate the users on how they can easily identify deepfakes. It would help in critical thinking and users will be skeptical while watching online content.
2. Traceability and Clarity: System to track and know the origin and manipulation of the content to hold the creators of such content accountable.
3. Participation with law enforcement: It would help the law enforcement agencies to a great extent if the platforms cooperate and partner with them. It would help those agencies in investigating and prosecuting the malicious use of deepfake.
4. Digital literacy programs will also help the citizens to understand how the deepfake content is different from original. It will help them to differentiate and not be misled by such content.
   In this way, it would be easier to align with the age of artificial intelligence and deepfakes. AI and its related cyber crime have created a new category of crime which was non-existent before. Therefore, necessary action must be taken by the government in regard to this.

Deepfakes are a dangerous advancement in the age of artificial intelligence, but they also carry real plusses in the realms of entertainment, education, and healthcare. But on the other hand, malpractices of these advances are causing a great threat to consent, reputation, privacy, and integrity of democracy. Lack of a particular legal regime in India has caused one to rely on some scattered but indirect provisions of law, which are generally inadequate in handling the particular damage caused by deepfakes. Internationally conducted comparison reveals that regulation, transparency, and accountability requirements can cause stagnation in such developments. Thus, a balanced regulatory framework for fundamental rights protection, enabling responsible innovation in this new techno-wave of artificial intelligence, is the need of the hour for India without delay, because improper use of deepfakes may badly create distrusts in the good promises of democracy and democracy of trust.

---

[1] Bharatiya Nyaya Sanhita 2023, s 358.

[2] Bharatiya Nyaya Sanhita 2023, s 271(2)(a) and s 75; Indecent Representation of Women (Prohibition) Act 1986.

[3] Constitution of India 1950, arts 19(1)(a) and 21.

[4] Bharatiya Nyaya Sanhita 2023, ss 111, 319 and 358; Information Technology Act 2000, ss 66C, 66D and 67; Information Technology (Intermediary Guidelines and Digital Media Ethics Code) Rules 2021, rr 3(1)(b)(viii) and 3(2)(b); Digital Personal Data Protection Act 2023, s 6.

[5] K.S. Puttaswamy v Union of India (2017) 10 SCC 1.

[6] Gaurav Bhatia v Naveen Kumar 2024 SCC OnLine Del 2704.

[7] National Stock Exchange of India Ltd v Meta Platforms Inc, Interim Application (L) No 21456 of 2024 in Comm IPR Suit (L) No 21111 of 2024 (Bombay HC, 16 July 2024).

[8] Karan Johar v Ashok Kumar/John Doe & Ors 2025 SCC OnLine Del 6108.

[9] Suniel Shetty v John Doe & Ors, Commercial IP Suit (L) No 32130/2025 (Bombay HC, 10 October 2025).

[10] WhatsApp LLC v Competition Commission of India 2022 SCC OnLine Del 2582.