Copyright applications in India have risen by 80.3 per cent, from 24,451 to 44,095, according to a five-year assessment of the country's intellectual property system.
The findings come from State of Intellectual Property in India, prepared by the Advanced Study Institute of Asia and published by Manohar. The report looks beyond the number of patents, trademarks and copyrights filed, granted and registered, to the institutions, licensing systems, litigation and commercial structures that determine whether those rights create value.
Revenue from copyright rose over the same period, from Rs 248.81 lakh in 2020-21 to Rs 364.76 lakh in 2024-25. Of the copyrights registered during 2024-25, literary works accounted for 65.1 per cent of the total.
The Copyright Office, like the Patent Office and the Trademarks Office, has a substantial backlog. In 2023-24 it examined 56,260 applications, disposed of 45,730 and granted 38,002 registrations. In 2024-25, 44,095 applications were filed and 26,767 disposed of, leaving 17,328 pending.
The report's central caution is about what those totals can and cannot show. "A filing, registration or grant is a legal event. It does not necessarily tell us whether an asset has been licensed, commercialised, enforced or converted into economic value."
Music illustrates why. In India, different rights in a musical work or sound recording may be owned and enforced by different entities involved in publishing, performing, recording and broadcasting, so a single digital music service can implicate rights held by several parties at once.
The courts have had to work through that complexity as the technology changed. In Tips Industries Ltd. v. Wynk Music Ltd., the Bombay High Court held that Section 31D of the Copyright Act, 1957 did not extend to internet broadcasting or online streaming, a position the court's Division Bench later affirmed on appeal.
The same question is now being asked of artificial intelligence, on a larger scale. In ANI Media Pvt. Ltd. v. OpenAI OpCo LLC, the Delhi High Court declined to grant an interim injunction in July 2026, and in September 2026 the Division Bench hearing the appeal also declined interim relief and sought OpenAI's response. The case is ongoing.
Alongside the litigation, the government has explored blanket or "no-fault" licensing, text and data mining exceptions, opt-out mechanisms and extended collective licensing. A discussion paper proposes a blanket licence for training, with a reconciliation fee payable on commercialisation.
Read together, the two findings at the centre of the report are harder to reconcile than they first appear. Filings rose 80.3 per cent. Revenue rose from Rs 248.81 lakh to Rs 364.76 lakh, which is about 47 per cent. The system is generating legal events faster than it is generating money.
This newspaper accepts the authors' framing in full and simply follows it. If a registration is a legal event rather than an asset, then the thing India has reliably increased production of is the legal event, and the body producing them is the Copyright Office.
Judged as a publisher, its figures are formidable. In 2023-24 it brought out 38,002 titles, and 65.1 per cent of its 2024-25 catalogue was literary.
Its slush pile is in proportion. Some 17,328 works are held unreleased at the end of 2024-25, each one submitted, received, and awaiting the decision that will turn it into property. Nothing about them has been rejected. They have simply not yet been granted the status their own existence is a request for.
The Department of Unnecessary Research has one outstanding question for the registry, and regards it as prior to all the others about artificial intelligence. The Copyright Office now produces tens of thousands of original literary determinations a year, and nobody has established who owns them, or where in the building they would be filed.