AI Deepfakes Causing Irreversible Harm Before Takedown: Justice PS Narasimha Flags Legal Gaps

Delivering the 5th Soli Sorabjee Memorial Lecture on the theme “Human Rights in the Empire of Artificial Intelligence,” Justice Narasimha observed that non-consensual synthetic imagery has emerged as a significant threat, perpetuating gender-based violence and causing deep psychological harm. He cautioned that the current legal response remains largely reactive, with remedial measures such as takedown actions often coming only after the damage has already been done.

“Non-consensual, synthetic imagery disproportionately targets women, perpetuating gender-based violence and inflicting acute psychological harm. Our present legal framework responds reactively, but takedown comes after the harm has already occurred,” he said.

The lecture, organised by the Commonwealth Lawyers Association in collaboration with Nivaaran, also saw the presence of Justice P. B. Varale, Additional Solicitor General Anil Kaushik, and Senior Advocates including Vikas Singh, Jaideep Gupta and Rajiv Dutta.

At the outset, Justice Narasimha paid tribute to the late Soli Sorabjee, recalling his legacy as a defender of civil liberties and constitutional values. He highlighted Sorabjee’s emphasis on institutional independence and his view of law as a public responsibility rather than merely a profession.

Turning to the broader implications of artificial intelligence, Justice Narasimha described AI as a form of power that is both diffused and less visible, yet deeply consequential. He noted that unlike traditional state power, AI operates across borders and restructures decision-making in ways that are not always transparent.

“AI does not displace power; it reorganises it in less visible but equally impactful ways,” he observed, adding that such systems often challenge traditional notions of informed consent.

The judge also flagged concerns over the use of AI in governance, particularly in welfare delivery systems. While acknowledging the efficiency gains, he warned that algorithm-driven processes could lead to exclusion, especially for vulnerable populations. He emphasised that welfare distribution is not merely administrative but integral to the realisation of socio-economic rights.

“Application of AI in welfare systems may result in exclusion. This has direct implications for human rights,” he said, stressing that technological optimisation must not come at the cost of inclusivity.

Justice Narasimha further underlined the need to extend principles of due process to algorithmic decision-making. He argued that individuals must have safeguards against opaque or automated systems that impact their rights and entitlements.

On privacy, he pointed out that the traditional understanding of privacy as mere non-disclosure is no longer sufficient in the age of AI. Instead, privacy must encompass control over personal data and the governance of complex data-processing infrastructures. Referring to the Digital Personal Data Protection Act, 2023, he noted that while it marks progress, it does not fully address emerging concerns such as data inference and large-scale algorithmic processing.

Concluding his address, Justice Narasimha stressed that institutions must ensure that the integration of AI strengthens, rather than dilutes, fundamental rights. He called for a framework grounded in inclusion, fairness, and accountability, warning that technological advancement must remain aligned with constitutional values.

“The integration of AI must reinforce the core principles of human rights, not undermine them,” he said.