The Transparency Paradox: Supreme Court Lets DPDP Act Hobble the RTI, While Govt Punts on AI Regulation

The Constitutional Tug-of-War: Privacy vs. Transparency

The Supreme Court of India has just handed data protection practitioners and constitutional lawyers a massive, complex knot to untangle. In a pivotal February 2026 hearing, the Apex Court declined to stay the operation of the Digital Personal Data Protection (DPDP) Act, 2023, specifically refusing to freeze provisions that severely dilute the Right to Information (RTI) Act, 2005. By referring the pleas to a larger bench and calling it a “complex and sensitive issue,” the Court has effectively allowed a new, restrictive information regime to solidify in practice.

For practicing lawyers, this isn't just academic constitutional theory—it is a fundamental shift in how we litigate transparency, advise data fiduciaries, and navigate the glaring regulatory vacuum surrounding Artificial Intelligence.

The Death of the Public Interest Override

To understand the gravity of the litigation, we must look at the mechanics of the amendment. Section 44(3) of the DPDP Act amends Section 8(1)(j) of the RTI Act.

Before the DPDP Act, Section 8(1)(j) exempted personal information from disclosure unless the Central Public Information Officer (CPIO) was satisfied that the larger public interest justified the disclosure. Crucially, it contained a powerful proviso: information which cannot be denied to the Parliament or a State Legislature shall not be denied to any person.

The DPDP Act has taken a sledgehammer to this nuance. It replaces the entire clause with a blanket exemption for information that relates to "personal information." The public interest test? Gone. The legislative parity proviso? Erased.

"By removing the public interest override, the DPDP Act transforms privacy from a shield for the citizen into a fortress for the State. CPIOs now have statutory cover to reject virtually any RTI application that contains a name, an employee record, or a beneficiary detail."

Practice Implication: For counsel representing journalists, NGOs, or whistleblowers, the traditional RTI route is now severely compromised. We will likely see a surge in writ petitions under Article 226, attempting to bypass the neutered RTI Act by invoking the fundamental right to know under Article 19(1)(a). Litigators must prepare to argue that statutory amendments cannot extinguish constitutional rights, pushing for a harmonious construction where the DPDP Act does not shield bureaucratic malfeasance.

The AI Regulation Vacuum: "Make Do With What We Have"

While the Supreme Court wrestles with the DPDP-RTI clash, the Ministry of Electronics and Information Technology (MeitY) has dropped another bombshell for tech lawyers: India will not see a bespoke AI law anytime soon.

The government’s stance is that existing frameworks—namely the DPDP Act and the Copyright Act, 1957—are sufficient to govern AI innovation. This "wait and watch" approach forces IP and technology lawyers to shoehorn 21st-century generative AI disputes into 20th-century statutes.

If you are advising an AI startup or a mid-sized tech firm, this regulatory punt creates immediate compliance headaches. How do you apply the DPDP Act to Large Language Models (LLMs)?

• The Right to Erasure vs. Immutable Tech: Section 9 of the DPDP Act guarantees the right to correction and erasure of personal data. But how do you erase a specific individual's data once it has been baked into the weights and parameters of a trained neural network? Tech lawyers are now debating the legal adequacy of "machine unlearning." If a user withdraws consent, advising a client to simply delete the database is no longer enough; you must assess if the AI model itself needs to be rolled back or retrained, a commercially devastating prospect.
• Copyright Infringement via Scraping: Without a specific AI Act outlining safe harbors for text and data mining (TDM), IP lawyers must rely on the fair dealing exceptions under Section 52 of the Copyright Act. Expect aggressive litigation from publishers against AI firms, arguing that scraping copyrighted material for training data constitutes reproduction without a license under Section 14.

The Phased Rollout Mirage

Compounding these issues is the phased rollout of the DPDP regime. Privacy commentators are accurately noting that startups and mid-sized firms are dangerously exposed. The illusion that "phased" means "delayed" is leading to critical compliance gaps.

The DPDP Act is not a check-box compliance exercise like the older SPDI Rules under the IT Act. It requires structural data mapping. If your client is scraping data to build an AI tool, they are a Data Fiduciary. They need verifiable consent, purpose limitation, and a mechanism for grievance redressal. The government's reliance on the DPDP Act to regulate AI means that Data Protection Boards will eventually scrutinize algorithmic training datasets for privacy breaches.

The Verdict for Practitioners

We are entering an era of aggressive statutory friction. The Supreme Court's refusal to stay the RTI amendment emboldens a culture of state secrecy, forcing litigators back to constitutional basics. Simultaneously, the lack of an AI-specific law forces corporate and IP lawyers to become highly creative, stretching the DPDP Act and the Copyright Act to their absolute limits.

Do not wait for the larger Supreme Court bench in 2026 to settle the privacy vs. transparency debate, and do not wait for an AI Act that isn't coming. Advise your clients to audit their data pipelines today. In the absence of clear regulatory guardrails, the companies that survive will be those whose legal counsel builds compliance architectures robust enough to withstand both privacy audits and copyright infringement suits.