The Government is Passing the Buck on AI Regulation: Why the DPDP-IP Collision is the Next Big Litigation Goldmine

The "No New AI Law" Cop-Out

If you have been holding your breath for a comprehensive, standalone Artificial Intelligence statute in India, you can finally exhale. In July 2026, the Ministry of Electronics and Information Technology (MeitY) made its stance unequivocally clear: the government prefers to rely on existing laws rather than draft an AI-specific code. This effectively means that the regulatory burden of the AI boom has just been dumped squarely onto the shoulders of the Indian judiciary and practicing lawyers.

For the tech-policy idealist, this is a regulatory vacuum. But for the pragmatic Indian litigator, it is a litigation goldmine. By declaring that the combination of the existing Intellectual Property framework and the Digital Personal Data Protection (DPDP) Act, 2023 is sufficient to govern AI, the State has essentially initiated a massive, uncoordinated collision between copyright law, data privacy, and public transparency.

Copyright Act vs. Machine Learning: The IP Squeeze

Without a specific AI regulation addressing text and data mining (TDM), Indian IP lawyers must retrofit the Copyright Act, 1957 to handle generative AI. When an AI model scrapes proprietary data to train its neural networks, is it an infringement under Section 51, or does it fall under the "fair dealing" exception for research under Section 52(1)(a)?

Because the government refuses to legislate a specific TDM exception—unlike the EU or Japan—every major AI scraping dispute will now require extensive trial evidence to prove whether the output is "transformative" under Indian jurisprudence. If you represent content creators, publishers, or media houses, the strategy is clear: sue for copyright infringement now, because the statutory shield for AI companies does not exist.

The DPDP Act as India’s De Facto AI Law: The "Machine Unlearning" Nightmare

While the IP regime fights over the output, the DPDP Act is about to wreak havoc on the input. Commentary from April and June 2026 highlights a massive compliance headache for AI companies operating in India, stemming directly from the consent and erasure frameworks of the DPDP Act.

Under Section 6 of the DPDP Act, consent must be free, specific, informed, unconditional, and unambiguous. More importantly, withdrawal of consent must be as frictionless as giving it. This brings us to the operational nightmare of Section 12: the Right to Erasure.

"When a Data Principal withdraws consent, the Data Fiduciary must erase their personal data. But how do you 'erase' data that has already been digested by a Large Language Model's neural network?"

The answer is a nascent technical concept called machine unlearning. For corporate and tech lawyers, this is where the rubber meets the road. You can no longer draft generic privacy policies and claim compliance. If you represent a Data Fiduciary deploying AI, your data processing agreements must now mandate technical capabilities for machine unlearning. If a user demands erasure of their personal data from your client's training dataset, failing to comply because "the algorithm already learned it" will not save your client from the Data Protection Board's draconian penalties.

The RTI Act Casualty: Section 8(1)(j) in the Supreme Court

While we grapple with AI, the DPDP Act is simultaneously cannibalizing transparency laws. On February 16, 2026, the Supreme Court referred challenges regarding the DPDP Act’s amendments to the Right to Information (RTI) Act, 2005 to a larger bench. Crucially, the Court did not stay the operation of the Act, noting only that there were "some creases to be ironed out."

For writ practitioners and civil rights lawyers, these are not mere "creases." Section 44(3) of the DPDP Act fundamentally alters Section 8(1)(j) of the RTI Act. Previously, personal information could be disclosed under the RTI Act if there was a larger public interest. The DPDP amendment deletes this public interest caveat, creating a near-blanket bar on the disclosure of any personal information by public authorities.

This is a severe blow to investigative journalism, anti-corruption litigation, and routine administrative law practice. If a lawyer files an RTI to uncover the beneficiaries of a corrupt government tender, the Public Information Officer (PIO) can now legally reject it, citing the un-amended blanket protection of "personal data" under the DPDP Rules 2025.

The Takeaway for Practitioners

The government's "wait and watch" approach to technology regulation has inadvertently made the DPDP Act the most powerful, and dangerous, piece of legislation for tech and civil practice in this decade. Here is what you need to change in your practice today:

1. For IP/Tech Lawyers: Stop waiting for an AI law. Start drafting TDM (Text and Data Mining) opt-out clauses for your media clients, and force AI developers to prove algorithmic compliance with the DPDP Act's right to erasure.

2. For Litigators: Prepare for a massive wave of writ petitions challenging PIO rejections under the amended RTI Act. Until the Supreme Court’s larger bench rules on the constitutionality of Section 44(3) of the DPDP Act, public authorities will use data privacy as a shield against accountability.

The State has abdicated its role in AI regulation. It is now up to the Bar to define the boundaries of algorithmic accountability through aggressive, creative litigation under existing laws.