“The Government is cognizant of the risks associated with new and evolving technologies including Artificial Intelligence (AI), which may generate misinformation through deep fakes,” said the Ministry of Electronics & Information Technology, when Members of Parliament in Rajya Sabha enquired about the government’s plan to tackle deep fakes created using artificial intelligence, on July 21, 2023.
MP Jawhar Sircar raised concerns about the use of deep fakes in the upcoming 2024 elections and also asked about the government’s plan to detect and regulate such technology. Separately, MP Surendra Singh Nagar asked about the government’s plan to tackle the transmission of such deep fakes on social media.
What is a deep fake? Deep fake refers to a hoax or fake content on the internet— images, audio, or video–created using artificial intelligence. These fake images or videos appear convincing enough to represent someone’s original voice, actions or even facial expressions. Recently, deep fakes of a number of famous personalities, including celebrities and politicians, are being used to spread tailored messages in order to spread misinformation or propaganda-related content on the internet. For example, in 2020, Manoj Tiwari, a Bharatiya Janata Party politician from Delhi, used deep fake videos during an election campaign. According to a report by Hindustan Times, the BJP’s Delhi unit used a communications company to produce at least two videos of the party’s leader Manoj Tiwari to show him speaking in Haryanvi and English by altering an original video of Tiwari, which was recorded in Hindi. Recently, Rest of the World also reported on how deep fakes can be used by Indian politicians either as a cover or to target political opponents during elections.
Key questions on deep fakes in the Parliament on July 21:
- What are the government’s plans to take on Artificial Intelligence-created deep fakes that deceive people by making someone appear to say or do something they did not do or say?
- Whether the government is satisfied with the tools designed to detect manipulated content capable of deep fakes or would it seek to cooperate with proven fact-checkers?
- Whether the government has examined the new regulations issued by other countries like China that aim to protect people from uninvited impersonation using deep synthesis technology?
- Whether the government would take adequate steps before the 2024 elections to stop or punish deep fake impersonation?
- Whether the government is aware that deep fake images are taking the form of a major social issue with potential implications for the integrity of the country?
- Whether the IT Ministry has any positive solution to stop the transmission of deep fake images on social media platforms? And the response of the social media interventions in this regard?
Why it matters: As India continues to shy away from deliberating on AI regulations, threats of risks associated with the use of such technologies are already being witnessed in processes like elections that can potentially impact democratic principles. Deep fakes can be used to deceive people, facilitate impersonation scams, extortion, and financial frauds, and essentially, deceive people. The United States Federal Trade Commission in March warned businesses of contributing to “AI deception” if such risks are not evaluated. We are yet to see any such similar statements by Indian regulators. Are existing IT Rules sufficient to address these threats and ensure accountability from companies involved in AI research and deployment of products?
Article continues below ⬇, you might also want to read:
- US Govt Warns Businesses Of ‘AI Deception’, Calls For Mitigating Risks Before Launching AI Products
- US Govt Warns Businesses To Keep A Check On Their AI Product Claims
- TRAI Recommends Setting Up Regulator For Artificial Intelligence In India
- Bard Now Available In Over 40 Languages Including 7 Indian Languages: Here’s What It Can And Cannot Do
Government cited IT Rules for protection from deep fakes:
Minister of State (MoS) for Electronics and Information Technology Rajeev Chandrasekhar responded to the queries stating that the Information Technology Act, 2000, and the associated IT Rules, 2021, provide safeguards from such threats related to deep fakes to citizens/users.
“These rules cast specific obligation on intermediaries, including social media intermediaries, to observe due diligence and provide that if they fail to observe such due diligence, they shall no longer be exempted from their liability under law for third-party information or data or communication link hosted by them,” Chandrasekhar added.
The MoS reiterated in both responses that Sub-clause (vi) of clause (b) of sub-rule (1) of Rule 3 under the IT Rules, obligates an intermediary including a social media intermediary to observe due diligence while discharging its duties and make reasonable efforts to ensure that a user must not “host, display, upload, modify, publish, transmit, store, update or share any information that impersonates another person or threatens the unity, integrity, defence, security or sovereignty of India or public order, or prevents investigation, or violates any law”. He also stated that Section 66D of the IT Act prohibits impersonation and cheating using any computer resource.
Additionally, the IT Rules also lay out provisions for an intermediary to act within 24 hours to remove or disable access to any artificially morphed content about an individual on receiving any such complaint. Individuals can also appeal online regarding such complaints to the Grievance Appellate Committees in case they are dissatisfied with the action taken by any platform.
“In addition, the Ministry of Home Affairs operates a National Cyber Crime Reporting Portal (www.cybercrime.gov.in) to enable citizens to report complaints pertaining to all types of cybercrimes, and also operates a toll-free helpline (1930),” the Minister added.
While the MoS listed out measures under the IT Rules for cybercrimes related to impersonation, it is unclear if deepfakes can be treated in the same manner as impersonation. Rajeev Chandrasekhar also did not clarify whether the government is working on methods to address problems related to election misinformation associated with such technologies.
STAY ON TOP OF TECH NEWS: Our daily newsletter with the top story of the day from MediaNama, delivered to your inbox before 9 AM. Click here to sign up today!