The Digital India Act (DIA) Draft 2026
The Digital India Act (DIA) Draft 2026
Context: The Ministry of Electronics and IT (MeitY) released the draft Digital India Bill, 2026 for public consultation in mid-January. Key Theme: From 'Safe Harbour' to 'Digital Accountability'. Keywords: Safe Harbour 2.0, Algorithmic Accountability, Digital India Authority, Do-No-Harm Principle.
1. The Context: Retiring the 26-Year-Old 'IT Act'
The existing Information Technology Act, 2000 was drafted when "Google" was a startup and "Facebook" didn't exist. It was designed to protect intermediaries (ISPs/websites) to let the internet grow.
In January 2026, the government officially acknowledged that the "Growth Phase" is over and the "Harm Phase" has begun. The draft DIA aims to replace the IT Act with a framework focused not on connecting people, but on protecting them.
2. The Core Shift: Safe Harbour 2.0
The most debated provision in the January draft is the dilution of Section 79 (Safe Harbour).
- Old Norm: Platforms like X/Instagram had "immunity by default." If a user posted hate speech, the platform was not liable unless they failed to remove it after a court order.
- The 2026 Draft: Safe Harbour is now "Earned, not Given."
- To claim immunity, "Significant Digital Intermediaries" (SDIs) must prove they took proactive measures to prevent specific harms (e.g., child abuse, deepfakes).
- Mains Implication: This shifts the burden of proof. It incentivizes platforms to use "Pre-emptive AI Moderation" rather than waiting for a takedown notice.
3. Algorithmic Accountability & 'Do-No-Harm'
For the first time, Indian law proposes to regulate the code itself.
- The Provision: The draft mandates that high-risk platforms must submit to "Algorithmic Audits" by the proposed Digital India Authority.
- The Logic: If an algorithm (e.g., YouTube Recommendations) amplifies polarization or self-harm content to increase "watch time," the platform can be penalized.
- The 'Do-No-Harm' Duty: The draft introduces a fiduciary duty on platforms towards "Digital Nagriks" (citizens). They must ensure their design does not cause psychological harm to children (e.g., addictive infinite scrolls).
4. Taming the 'Deepfake' Dragon
In response to the 2025 deepfake crisis, the DIA draft creates specific offences for Synthetic Media.
- Labelling Mandate: Any content generated by AI must have a visible and metadata-based "Origin Label."
- Criminal Liability: Creating a deepfake with "malicious intent" (to defame, incite violence, or defraud) is proposed as a non-bailable offence, distinct from standard forgery.
- The 'Platform's Duty': If a deepfake goes viral, the platform cannot hide behind "we didn't know." They must show they had technologies in place to detect synthetic patterns.
5. Mains Analysis: The "State vs. Big Tech" Battle
- The Sovereignty Argument: The DIA asserts "Digital Sovereignty." It rejects the view that the internet is a borderless global commons. It treats the Indian internet as a sovereign space where Indian laws (and not Silicon Valley "Community Guidelines") are supreme.
- The "Chilling Effect" Critique: Critics (Internet Freedom Foundation, etc.) argue that "Earned Safe Harbour" will force platforms to over-censor. To avoid liability, platforms might proactively delete even legitimate political speech, effectively becoming "Private Censors" for the State.
