You can access the original order from here
The Calcutta High Court on May 20 refused to direct OpenAI to surface IndiaMart links in ChatGPT responses. Justice Ravi Krishan Kapur dismissed IndiaMart InterMesh Ltd’s interim relief application, finding no prima facie case and holding that the balance of convenience against granting any order. This reverses the court’s December 2025 order, where the same judge found a “strong prima facie case” of selective discrimination and said IndiaMart was excluded “without any logic.” After hearing OpenAI, that view did not hold.
Why it matters: India has 100 million weekly ChatGPT users, the second largest base globally. This is the first Indian court ruling on whether a business has legal recourse when an AI platform silently excludes it. The answer, for now, is no.
“AI systems are not merely ranking links. They are actively constructing the consumer’s informational universe,” said Navneet Sharma, VB Padode Chair Professor of Competition Policy at Vijaybhoomi University.
“Once AI assistants become the primary discovery interface for products, suppliers, and commercial information, exclusion from their outputs can materially affect market access. The issue is no longer merely ‘speech’ or ‘search’; it is one of platform neutrality and competitive neutrality,” Sharma added.
IndiaMart’s allegations:
OpenAI’s defence: OpenAI argued IndiaMart had no legal right under contract, statute, or constitutional law to demand visibility on ChatGPT. Its exclusion decision relied on the USTR’s 2024 Notorious Markets List, an annual American trade watchlist that flags markets alleged to facilitate intellectual property (IP) violations. It carries no legal force in India, and listed entities receive no notice or opportunity to respond before being named.
Sharma called this a digital sovereignty concern: “If global AI platforms operationalise foreign governmental risk classifications into discoverability rules, they effectively become private instruments of extraterritorial regulatory influence. That raises concerns of competitive neutrality and procedural fairness for Indian businesses.
“Particularly troubling is the allegation that similarly situated foreign marketplaces allegedly remained visible while IndiaMart was excluded. If true, that moves the issue from content moderation into potential discriminatory market treatment.”
The court also noted IndiaMart had previously blocked ChatGPT from crawling its website, and used that against the company.
Court’s findings
The gatekeeper question: The court’s refusal to grant relief does not resolve the deeper question of whether AI platforms function as gatekeepers controlling market access.
Sharma laid out a four-point test: “A gatekeeper concern arises when the platform controls visibility, the criteria are opaque, the exclusion materially affects downstream commerce, and affected businesses lack procedural safeguards or meaningful appeal mechanisms.” All four conditions, he argued, apply to ChatGPT’s treatment of IndiaMart.
Consumer protection gap: A user asking ChatGPT where to source a product has no idea IndiaMart has been excluded or why. “Conversational AI creates an illusion of completeness and objectivity. A user asking ‘Where can I source industrial bearings?’ is unlikely to know which marketplaces are excluded, whether exclusions are policy-driven, or whether commercial or reputational filters are being applied. Unlike search engines, AI systems often do not expose the underlying index, ranking logic, or omitted alternatives. That asymmetry creates a serious transparency deficit,” Sharma said.
Under the Consumer Protection Act, 2019, he argued, this raises concerns around unfair trade practices and misleading representations by omission: “The issue is not only deception by commission, but also material omission. If consumers reasonably assume AI outputs are comprehensive or neutral and commercial exclusion decisions are undisclosed, such exclusions distort purchasing choices.”
Intermediary vs originator: The most consequential part of the order for Indian AI governance is the court’s treatment of ChatGPT’s legal status under the Information Technology (IT) Act, 2000. An intermediary under the IT Act is a platform that passively hosts or transmits content created by others, like Google or WhatsApp. Intermediaries get safe harbour protection but are also bound by IT Rules, 2021 obligations around transparency, grievance redressal, and content takedowns. An originator is a content creator, not a passive conduit, and carries no such obligations.
IndiaMart argued ChatGPT qualifies as an intermediary, which would give it a regulatory hook. Justice Kapur distinguished conventional search engines, which crawl and rank existing content, from large language models (LLMs), which synthesise and generate new responses. The court observed prima facie that ChatGPT may be closer to an originator, but said this requires fuller adjudication with technical and expert evidence at trial.
Who has jurisdiction: Until the intermediary vs originator question is settled, no single body has a clear mandate. Sharma outlined the options:
The regulatory gap: “India is entering a major regulatory gap. Existing frameworks were designed for search engines, marketplaces, social media intermediaries, or telecom-style access infrastructure. Generative AI combines all four functions simultaneously: search, recommendation, curation, and synthesis,” Sharma said. Currently there is:
The IndiaMart contradiction: As MediaNama reported in Q4FY26 earnings coverage, IndiaMart’s management told investors ChatGPT is “not yet very good at B2B queries” with no material impact on inquiries. The company simultaneously argued in court that AI-driven exclusion caused significant commercial harm. The court’s dismissal does not end the main suit, but IndiaMart’s credibility on the harm question has taken a hit.
Broader pattern: The IndiaMART matter, Sharma said, “may become one of India’s earliest cases testing whether generative AI platforms should bear obligations analogous to search neutrality, platform neutrality, and procedural fairness duties traditionally debated in digital competition law.”
AI platforms replacing search as India’s primary discovery layer is a pattern MediaNama has tracked from MakeMyTrip’s Myra raising self-preferencing concerns in August 2025 to Google’s agentic commerce protocol in January 2026.
In each case the same gap emerges: AI platforms make visibility decisions affecting Indian businesses and consumers with no Indian law requiring them to explain those decisions.
The parties are directed to take steps for expeditious hearing of the main suit.
Also read:
For You
此内容由惯性聚合(RSS阅读器)自动聚合整理，仅供阅读参考。 原文来自 — 版权归原作者所有。