E-Paper
In focus at India AI Summit: Big tech is feasting on news and refusing to pay for it, say publishers
News businesses, legacy and new, face a structural shock as AI platforms use their original work to train models and cut traffic without sharing value.
One glimpse at the long queues and teeming corridors at New Delhi’s Bharat Mandapam - the venue for the IndiaAI Impact Summit - is enough to show how democratised the generative AI boom has become. While a large part of this phenomenon is being built on journalism, big tech companies continue to undermine the economics of news publishing, experts warned at the summit.
News businesses - both legacy organisations and upstarts - are facing a structural shock that has become starkly visible over the past year: major AI platforms are consuming their originally produced work to train models and affecting user traffic, all without sharing value.
“AI summaries in search have led to as many as 60% of searches no longer going to websites. How are we funding journalism? There have been huge falls in the past 12 months,” said Robert Whitehead, Digital Platform Initiatives Lead at the International News Media Association. He was speaking on the opening day of the IndiaAI Impact Summit 2026, where leaders of the Digital News Publishers Association stressed that AI platforms were extracting value from publishers without fair compensation.
“This is already depriving revenue from the businesses that are funding the accuracy of the data that is needed for a sovereign model or any other large language model,” Whitehead said.
Track live updates on the AI Summit here.
Tech giants vs news organisations: The way forward
Business models of digital news publishers rely significantly on traffic from search and social platforms. At the panel titled ‘AI and Media: Opportunities, Responsible Pathways, and the Road Ahead’, speakers said that the rise of AI-generated answers and summaries is accelerating the “zero-click” trend, in which users obtain information without visiting source sites.
At the same time, AI models depend heavily on news publishers for training and outputs, often without revenue-sharing or attribution. Experts say there’s a growing imbalance in how AI platforms are capturing audience and value, while newsrooms bear the cost of producing credible information.
“I feel that we don’t get the same exceptions from the tech companies like American media brands and I see that as digital imperialism,” said Kalli Purie, Vice Chairperson and Executive Editor-in-Chief, India Today Group.
AI models require high-quality, verified text, much of which comes from newsrooms that fund original reporting. “We are the guys who are funding the reporters to go out and break the original story… We are not getting credited for it. We are not getting any payment for it,” she added.
Global analytics firms and publishers have reported sharp declines in referral traffic from search, even as the presence of AI overviews and answer boxes expands. Industry-wide estimates cited by media executives indicate there are 20–60% fewer click-throughs for news-related queries.
“Over a period of time, how will we sustain reporters,” asked Purie. “If we don’t have original stories, AI will eat AI… it’ll keep making stories from synthetic content.”
How should AI treat news?
Executives at the panel also discussed how journalism must be treated differently from other forms of content in AI systems. News carries consequences far beyond engagement metrics, and it can influence elections, markets, social stability, and national security, they said.
“The journalistic content is not like free-floating content of the internet. It is something which is intellectual property. It gets created with investment, with infrastructure, with talent,” said Mohit Jain, Chief Operating Officer and Executive Director, Bennett Coleman & Company Limited.
He argued that when AI tools summarise or redistribute news, they influence public discourse in ways comparable to publishers themselves. “Anything that begins to participate in a democratic process deserves a different standard of care.”
Watch: Full session on AI and Media from India AI Summit at Bharat Mandapam, New Delhi
India's pitch to host AI infra
The Indian government has pitched the IndiaAI Impact Summit as a global opportunity to build AI governance and geopolitical consensus. This is the first time the event is being organised in a developing country.
“The theme of the summit is... welfare for all, happiness for all, reflecting our shared commitment to harnessing Artificial Intelligence for human-centric progress,” Prime Minister Narendra Modi said. Among the main speakers are Alphabet CEO Sundar Pichai, OpenAI CEO Sam Altman, Anthropic CEO Dario Amodei, Reliance Chairman Mukesh Ambani, and Google DeepMind CEO Demis Hassabis.
Also read: Over 600 startups, 70,000 sq-m area, 2.5 lakh visitors - Mega scale of global AI summit in Delhi
At a panel discussion on Monday, Sujata Gupta, Secretary General, DNPA, underscored the significance of journalism in the AI era. As India builds its AI capabilities, accountability, attribution, and institutional trust must remain foundational, she said.
Beyond revenue and business impact, panelists also emphasised that models primarily built on foreign-language datasets often underperform in Indian languages and fail to capture regional nuance. “Tier-2 and Tier-3 India form the backbone of the country’s demographic strength, and AI systems must reflect linguistic diversity and cultural context,” said Pawan Agarwal, Deputy Managing Director, Dainik Bhaskar Group.
“If not protected now… you will only see news generated from users and there will be no newsrooms,” he added. “There will be nobody taking accountability.”
Others argued that the ultimate impact would fall on the citizens’ information ecosystem. “The biggest loss will be to the consumer,” said Tanmay Maheshwari, Managing Director, Amar Ujala Group. “You cannot put me into an information bubble all the time.”
Big tech companies generally rejected claims that AI is undermining journalism. They say their platforms send billions of clicks annually to news sites and that AI summaries often promote publisher content rather than replace it.
Also read: IT, BPO services will disappear in five years, says tech billionaire Vinod Khosla
International developments formed a significant part of the discussion, including the European Union’s AI Act, which introduces labelling obligations for AI-generated content; moves by France and Germany to operationalise rights frameworks requiring technology platforms to negotiate compensation agreements with publishers; and Australia’s News Media Bargaining Code, which mandates commercial negotiations between platforms and publishers and has led to substantial remuneration deals.
The discussion also addressed opportunities arising from the AI revolution. Experts highlighted how AI can deepen archives, enhance contextual journalism, improve newsroom efficiency, and strengthen subscription models. “The challenge is to ensure that AI enhances long-term credibility rather than eroding it,” said Navaneeth LV, Chief Executive Officer, The Hindu Group.
