India’s New AI Accountability Bill Would Force Global Platforms to Audit Algorithms Used on Indian Users

When a billion-plus people scroll, search, and speak through platforms designed elsewhere, who answers for what those platforms decide to show them? India is now demanding an answer in writing — and on a deadline.

India’s Ministry of Electronics and Information Technology (MEITY) has circulated a draft AI Accountability Bill that would require any digital platform serving more than five million Indian users to submit annual third-party audits of their recommendation systems and content-moderation algorithms. The proposal targets the core machinery of platforms like Google, Meta, and OpenAI — not just their stated policies, but the automated systems that shape what 1.4 billion people see, read, and believe.

If enacted in its current form, the bill could establish a regulatory benchmark more operationally demanding than the EU AI Act, and one with far greater demographic reach.

—

What the Draft Bill Actually Requires

At its center, the bill mandates structured transparency from any platform that crosses the five-million-user threshold — a bar that virtually every major global technology company already clears in India.

Under the draft framework, covered platforms would be required to commission independent, third-party audits of their algorithms on an annual basis. Those audits would examine recommendation engines and content-moderation systems for bias, discriminatory outputs, and systemic risks to public discourse. Platforms would submit audit findings to a designated regulatory authority under MEITY and make summary disclosures available to the public.

The bill also contemplates a tiered risk-classification system for AI applications, placing heightened scrutiny on systems that influence political content, financial decisions, and health information — categories where algorithmic errors carry outsized social consequences.

Critically, the draft does not permit platforms to self-certify. It requires external auditors, creating a new professional and institutional infrastructure around AI governance that India does not yet fully possess.

—

Why This Goes Further Than the EU AI Act

The EU AI Act, which entered into force in 2024, is widely regarded as the world’s most comprehensive AI regulation. It classifies AI systems by risk level and imposes conformity assessments, transparency obligations, and prohibitions on certain high-risk applications.

But the EU framework has notable gaps when it comes to recommendation algorithms. Social media feeds, search rankings, and content-moderation systems — the tools that most directly shape public opinion at scale — occupy a regulatory gray zone under the EU’s current approach, addressed partly through the Digital Services Act rather than the AI Act itself.

India’s draft bill, by contrast, places recommendation and moderation algorithms explicitly at the center of its platform regulation mandate. For AI governance observers, this is a significant architectural choice. It signals that MEITY views algorithmic curation not as a peripheral product feature but as a core infrastructure of public life requiring direct accountability.

The scale of India’s user base amplifies the stakes considerably. Regulatory requirements that apply to Indian users effectively bear on the global product decisions of every major platform, given the cost and complexity of maintaining jurisdiction-specific algorithmic systems.

—

The Compliance Challenge for Global Platforms

For Google, Meta, OpenAI, and their peers, the bill presents a compliance architecture unlike anything currently required in their largest markets.

Annual algorithm audits are not a trivial undertaking. Recommendation systems are often proprietary, deeply integrated, and continuously updated through machine learning processes that make static snapshots difficult to interpret meaningfully. Defining what an audit should measure — and what a satisfactory result looks like — requires technical standards that do not yet exist at the level of specificity the bill envisions.

Platforms will likely argue that substantive algorithm audits require access to training data, model weights, and internal performance metrics they consider core intellectual property. Negotiating the boundary between regulatory transparency and trade secret protection will be among the bill’s most contested implementation questions.

Auditor capacity presents a separate challenge. India’s technology audit ecosystem, while growing, has not historically operated at the scale or technical depth that AI system audits demand. Building that capacity — through accreditation frameworks, technical standards bodies, and professional training — is a prerequisite for the bill’s ambitions to be realized rather than merely performed.

—

What Digital Rights Advocates Are Watching

For Indian digital rights advocates, the bill represents both a genuine opportunity and a tangible risk.

The opportunity is real accountability for systems that have demonstrably shaped political discourse, amplified misinformation, and produced discriminatory content-moderation outcomes affecting marginalized communities. Mandatory third-party audits, if designed with civil society input and robust public disclosure requirements, could give researchers, journalists, and affected communities meaningful tools to challenge platform behavior.

The risk lies in implementation. Regulatory frameworks built around government access to platform data can be repurposed for surveillance or political control. Advocates are pressing for audit processes that are independent not only from platforms but from government influence — with findings made available to the public rather than held exclusively by MEITY.

The bill’s treatment of audit confidentiality, whistleblower protections for auditors, and civil society participation in standard-setting will ultimately determine whether it functions as a genuine accountability mechanism or an exercise in compliance theater.

—

A Precedent With Global Consequences

India has a history of setting regulatory directions that other large democracies follow, adapt, or react to. Its data protection framework, its approach to platform intermediary liability, and its digital public infrastructure model have all drawn sustained international attention.

The AI Accountability Bill arrives at a moment when governments worldwide are searching for workable models of AI governance that move beyond principles and produce enforceable obligations. A bill that requires annual algorithm audits from the world’s largest platforms — and builds the institutional infrastructure to conduct them — would represent a meaningful step beyond the aspirational language that dominates most national AI strategies.

The draft remains in circulation. Stakeholder consultations, parliamentary scrutiny, and industry negotiation all lie ahead. But the direction MEITY has signaled is unambiguous: in India’s regulatory vision, algorithmic power over public life is not a private matter. It is a public accountability question — and the audit clock is already running.