As India’s IT majors race to badge themselves “AI-first,” claiming triple-digit patent counts, 190 here, 400 there, have been the calling cards. But do those filings translate into real innovation and revenue, or are they optics more than outcomes?
In the April 18, 2025, case of Recentive Analytics, Inc vs Fox Corp, the US Federal Circuit held that patent claims for applying generic machine learning to new environments such as event scheduling and network maps are ineligible for protection.
The court affirmed dismissal, finding no technological improvement or “inventive concept” beyond the abstract idea of using ML for those purposes.
Claims that simply recite applying a known ML process to a different dataset, without a technical innovation in the ML itself, do not become patent-eligible.
Against that backdrop, India’s AI patent filing boom looks impressive but complex.
Between 2010 and 2025, India saw about 86,000 AI-related patents filed, more than 25% of all technology patents filed in the country; generative AI accounts for 28% of India’s AI filings, according to a Nasscom report.
Yet, the grant ratio for AI patents is just 0.37% in India, markedly below that in the US and China.
Overall, the grant ratio for AI patents in India is 21.5%. For academia, it is only 1.1% while corporates see a higher 44.1%.
Leading the patent race
India’s IT companies have been aggressively filing for AI patents, with several seeing a steady stream of grants. These filings cover areas ranging from automation and analytics to generative AI, and companies highlight them as evidence of their ability to commercialise emerging technologies at scale. The focus has been on building large patent portfolios that showcase AI-led innovation.
Hansa Iyengar, practice leader (BFS & IT Services) at HFS Research, said patent announcements are often “crafted to impress clients and investors.”
Asked whether these filings reflect genuine innovation or simply serve as marketing signals, she said that the real test is whether those patents are used in products and deployments with measurable results.
So far, she observed, there is little evidence that Indian IT firms are earning meaningful licensing revenues from AI patents; most filings “mainly serve as signals of innovation capability rather than cash generators.”
That raises the practical question of how to separate patent quantity from patent quality.
According to Iyengar, patent counts do not equal innovation. More reliable signals include forward citations, international family size, renewals, and whether claims demonstrate a technical effect under India’s strict Section 3(k) of the Indian Patents Act, 1970, which excludes “a mathematical or business method or a computer programme per se or algorithms” from the definition of an invention.
Sachin Jain, principal director, LTIMindtree Research, said in Nasscom’s report that to achieve appropriate protection outcomes, filers must ensure clarity in articulation and conduct comprehensive prior research focused on novelty, usefulness and non-obviousness.
“The pace of innovation can be accelerated by fostering collaboration within the ecosystem, further simplifying and streamlining the process, incorporating technology-enabled reviews, implementing recognition programs, and enhancing IP protection awareness,” he said.
An Intellectual Property head of a leading telecom MNC was quoted in the report as saying that while the Indian patent office has made impressive improvements, lack of consistency in examination remains a challenge with some examiners accepting specifications that others reject.
He said this needs to be addressed and called for harmonisation across jurisdictions, especially between the United States Patent and Trademark Office, European Patent Office, and the Indian patent office.
According to the report, LTIMindtree drives IP-led innovation with its Idea2Impact framework, which manages the IP lifecycle and aligns R&D with IP strategy across AI, quantum, cybersecurity and other advanced technologies.
TCS builds IP awareness through its annual IP Week, using knowledge sessions and discussions to strengthen a culture of innovation.
Accenture, HCLTech, SAP Labs India, Birlasoft and Nutanix did not respond to AIM’s queries.
Nasscom’s head of AI, Ankit Bose said that an AI patent’s true value lies not in the filing, but in its impact.
“We champion a credibility standard built on three pillars: rigorous research, global peer validation, and demonstrable business outcomes, ensuring India builds a portfolio of genuine, market-ready innovations, not just numbers.”
The post AI Patents in Indian IT: Real Innovation or Just Theatrics? appeared first on Analytics India Magazine.