Democrat Jamie Raskin's confrontation with Republican colleagues over AI companies' use of copyrighted material without consent has exposed the deepest fault line in global tech regulation. For India — the world's largest IT services exporter and a fast-growing AI economy — the outcome will determine whether generative AI tools its workforce relies on are built on legally sound or legally toxic foundations.

Three words cut through the procedural fog of a US Congressional hearing like a machete: You stole our work. Congressman Jamie Raskin, the Maryland Democrat who has made a career of turning committee rooms into courtrooms, was not directing his fury at a foreign adversary or a corrupt contractor. He was staring down the single most powerful industry on the planet — artificial intelligence — and accusing it, in plain language, of building its fortune on the unpaid labour of every writer, artist, musician, photographer, and coder whose work was vacuumed into training datasets without so much as a notification, let alone a cheque.

The confrontation, as reported by The Times of India, has since ricocheted far beyond Capitol Hill. And the shockwave that should concern Indians most is not the spectacle — it is the substance. Because when the United States ultimately decides whether AI training on copyrighted material is legal or criminal, the ruling will not stay inside American borders. It will land, with the force of a regulatory earthquake, on every IT services company in Bengaluru, every freelance content creator in Mumbai, every AI startup in Hyderabad, and every young coder in Pune who is betting their career on generative tools whose legal foundations may turn out to be quicksand.

This is not hyperbole. This is arithmetic.

What Actually Happened in That Hearing Room

The Congressional hearing, part of the US legislature's intensifying scrutiny of AI's societal impact, brought together lawmakers from both parties, AI industry representatives, and advocates for creative workers. Raskin's intervention was the moment that turned a policy discussion into a viral confrontation. His argument, as reported by multiple outlets including The Times of India and corroborated by C-SPAN footage circulating online, was blunt: generative AI companies — from OpenAI to Meta to Google — ingested billions of pages of copyrighted books, articles, code repositories, photographs, and music tracks to train their large language models. The creators of that work were never asked. They were never paid. And now the companies built on their labour are valued in the hundreds of billions.

Republican members pushed back, framing the debate as innovation versus regulation — the familiar American political theatre of progress against protection. But what made Raskin's salvo land harder than a typical partisan exchange was the bipartisan unease visible in the room. According to reports, even some Republican members acknowledged that the status quo — where AI firms claim 'fair use' as a blanket defence for industrial-scale data ingestion — is legally untested and ethically uncomfortable.

The India Equation No One Is Talking About

Here is the dimension the American coverage largely missed, and the one India Herald's read of this story centres on: India is not a spectator in this fight. It is, arguably, the country with the most at stake.

Consider the numbers. India's IT and business process management industry generated an estimated $254 billion in revenue in FY2024, according to NASSCOM. A significant and rapidly growing share of that revenue now involves AI-adjacent services — building, fine-tuning, deploying, and maintaining generative AI systems for global clients. Indian IT giants like TCS, Infosys, Wipro, and HCLTech have publicly pivoted their strategies toward generative AI integration. Their client pitches, as reported in the Economic Times, increasingly hinge on the promise that Indian engineers can build, customise, and scale AI tools faster and cheaper than anyone else.

But what happens to that promise if the foundational models those tools are built on — GPT, Gemini, LLaMA, and their successors — are ruled to have been trained on stolen intellectual property? The legal exposure does not stay with OpenAI or Google. It cascades. Every downstream application, every enterprise deployment, every client solution built on a legally compromised model inherits the liability.

This is not theoretical. The New York Times, Getty Images, and dozens of individual creators have already filed copyright lawsuits against AI companies in US courts. If the US Congress legislates — or if courts rule — that AI training on copyrighted material requires consent and compensation, the cost of AI models will rise dramatically. Licensing fees, royalty structures, and compliance burdens will be baked into every API call. And those costs will flow directly into the contracts that Indian IT firms sign with their American and European clients.

Political Pulse

The backstage talk in Indian policy corridors, according to sources familiar with MEITY's AI governance consultations, is quieter but no less anxious. India's own copyright law — the Copyright Act of 1957, amended most recently in 2012 — has no explicit provision for AI training. The question of whether feeding copyrighted Indian-language content into a model constitutes infringement is, legally, an open void. The whisper in South Block, per sources tracking the India AI Mission's regulatory track, is that New Delhi is watching the US outcome closely before committing to its own framework — a classic wait-and-watch that could leave Indian creators unprotected for years while their work is ingested at scale.

There is a deeper political calculation here that the Raskin hearing has accidentally spotlit. India's ruling dispensation has been eager to position the country as a global AI powerhouse — the India AI Mission, launched with considerable fanfare, targets making India a top-three AI economy. But that ambition sits in direct tension with the copyright question. If India sides with creators and mandates consent-based training, it raises costs for its own AI startups and potentially slows the very ecosystem it is trying to build. If it sides with the AI companies and permits unrestricted training, it alienates a massive creative workforce — writers, musicians, filmmakers, photographers, coders — who are already feeling the squeeze of AI-generated content flooding their markets.

The political math is uncomfortable. The creative industries employ tens of millions and are culturally vocal. The IT industry employs fewer but generates vastly more export revenue. Which constituency does a government protect when their interests collide head-on? The Raskin hearing, for all its American theatrics, has forced that question into the open — and Indian policymakers cannot pretend it does not apply here.

[EMBED-SUGGESTION:tweet]

The Freelancer on the Frontline

Zoom out from the policy suites and the question becomes viscerally personal. India is home to the world's second-largest freelance workforce, according to a 2024 report cited by Mint. Millions of Indian freelancers — graphic designers, content writers, translators, voice artists, coders — sell their services on global platforms. Generative AI, trained on work that looks and reads remarkably like theirs, is already undercutting their rates. A Fiverr logo design that once fetched $50 now competes with a Midjourney output that costs the client $0.02 in compute.

If the US Congress affirms that AI companies must compensate creators whose work was used for training, it establishes a precedent — however imperfect — that human creative labour has legal value even in the age of machines. If Congress blinks, or if the courts carve out an expansive fair-use exemption for AI, it sends the opposite signal: your work is raw material, and raw material does not negotiate.

For millions of Indian workers, that is not a philosophical debate. It is a pay cheque.

What India Herald Sees Coming Next

India Herald's assessment, reading the trajectory of both the US legislative calendar and India's own regulatory signals, is this: the Raskin confrontation is the opening act, not the climax. The US is unlikely to pass comprehensive AI copyright legislation before the 2026 midterms — the issue is too commercially explosive and too politically cross-cutting. What is more likely, based on the pattern of recent Congressional hearings, is a series of executive orders and agency guidances (from the US Copyright Office and the FTC) that begin to narrow the fair-use defence for AI training without fully resolving it.

For India, the window of ambiguity is both an opportunity and a trap. The opportunity: India could move first, crafting a consent-and-compensation framework for AI training that becomes a global model — the way GDPR made the EU the de facto standard-setter for data privacy. The trap: if India waits too long, its creators' work will have already been ingested into models that are functionally impossible to un-train, and the leverage will have evaporated.

Watch for three signals in the months ahead. First, any movement by MEITY or the Department for Promotion of Industry and Internal Trade (DPIIT) on AI-specific copyright amendments. Second, the outcome of pending US court cases — particularly The New York Times v. OpenAI — which will set the legal baseline the world references. Third, whether Indian creative guilds and unions (the Film Writers Association, the Indian Newspaper Society, photography collectives) begin filing their own suits or lobbying campaigns. The moment Indian creators organise around this issue, the political calculus in Delhi shifts overnight.

Key Takeaways

1. Jamie Raskin's confrontation in the US Congress over AI copyright is not American political theatre — it is the opening salvo of a global regulatory war whose outcome will directly shape India's $254-billion IT industry and its millions of creative workers.

2. India's copyright law has no explicit provision for AI training, leaving creators legally unprotected while their work is ingested into models at industrial scale — a gap that policymakers are reportedly watching the US to fill, rather than acting independently.

3. The political tension in India is between protecting a culturally vocal creative workforce and nurturing an export-revenue-generating AI ecosystem — the Raskin hearing has forced that contradiction into the open, and the government will eventually have to pick a side.

Key Takeaways

• Jamie Raskin's US Congressional confrontation over AI copyright is the frontline of a global regulatory war with direct consequences for India's $254-billion IT industry.
• India's Copyright Act has no explicit provision for AI model training — a legal void that leaves millions of Indian creators unprotected.
• The political tension between India's creative workforce and its AI-growth ambitions will force policymakers to choose a side, and the US outcome will set the baseline.
• Indian freelancers — the world's second-largest freelance workforce — face existential wage pressure from AI tools trained on work indistinguishable from their own.
• Three signals to watch: MEITY copyright amendments, the NYT v. OpenAI ruling, and whether Indian creative guilds begin organising around AI training rights.

Frequently Asked Questions

What did Jamie Raskin say during the AI Congressional hearing?

Congressman Jamie Raskin accused AI companies of stealing creative workers' copyrighted material — books, articles, code, images, and music — to train their large language models without consent or compensation, calling it systematic theft during a heated Congressional hearing.

How does the US AI copyright debate affect Indian IT companies?

Indian IT giants like TCS, Infosys, and Wipro are pivoting toward generative AI services. If US law rules that AI models were trained on stolen IP, the legal liability and licensing costs cascade into every downstream application — including those built and deployed by Indian firms for global clients.

Does Indian copyright law cover AI training on copyrighted material?

No. India's Copyright Act of 1957, last amended in 2012, has no explicit provision addressing whether feeding copyrighted content into AI training datasets constitutes infringement — a legal void that leaves Indian creators unprotected.

What should Indian creators and coders watch for next?

Three key signals: any movement by MEITY or DPIIT on AI-specific copyright amendments, the outcome of US cases like The New York Times v. OpenAI, and whether Indian creative guilds begin organising legal or lobbying campaigns around AI training rights.