New Delhi, June 04, 2025: With generative AI emerging as a transformative force across the global economy, the Esya Centre and the AI Knowledge Consortium convened leading voices in a closed-door discussion. A highlight of the convening was the launch of a new report by the Esya Centre, by BJP Rajya Sabha MP Shri Sujeet Kumar.Titled “Rethinking Data and Competition: A Critical Assessment of the Data-Driven Market Tipping Theory,”the report challenges the notion that data accumulation necessarily leads to monopolistic outcomes and critiques various ex-ante regulatory measures proposed in India, such as forced data sharing and data cross-use restrictions.
Key findings from the report include:
- Data-driven firms often face diminishing returns on data aggregation – i.e., as firms collect more data, the incremental value of additional data often decreases.
- Big Data is fundamentally non-rivalrous, meaning that its use by one firm does not prevent another firm from collecting or using the same data. This non-rivalrousness can also have pro-competitive effects, since it allows smaller businesses to enter data-intensive markets and leverage innovative business models.
- Multi-homing undermines the exclusivity of data-driven network effects by enabling users to share the same data (e.g. browsing behaviour) across different providers. Research from India has found that users multi-home both on e-commerce platforms as well as AI models.
- No market requires generic Big Data – each market depends on specific data inputs tailored to its needs.
- Indian policymakers have proposed 3 kinds of ex-ante measures based on the data-driven market-tipping theory — data cross-use and combination restrictions under the Draft Digital Competition Bill 2024, forced data sharing under the Draft National E-Commerce Policy 2019, and data portability under the Draft Digital Competition Bill. These may jeopardise efficiency gains from data aggregation as well as consumer welfare. They may also createnormative tensions with intellectual property, data protection and security.
speaking on the report’s release, Meghna Bal, Co-Author of the report and Director at the Esya Centre, said,“India stands at a unique inflection point. As we shape a competitive and inclusive AI economy, our approach to data and competition must reflect the realities of digital ecosystems. This means moving beyond rigid data-sharing mandates and embracing frameworks that support voluntary sharing, alternative data sources like synthetic datasets, and better implementation of open government data policies. The report is a call to rethink legacy assumptions and design agile, innovation-friendly governance tailored to India’s needs.”
The report recommends alternatives such as supporting voluntary data sharing frameworks, improving open government data implementation, exploring alternative data sources like synthetic or open-source datasets, and introducing legal reforms to enable AI training.
Enabling Innovation Through Agile Governance
Speakers at the convening underscored that India’s regulatory approach must not merely respond to risk but actively enable innovation. While acknowledging the need for legal clarity in areas like AI safety and data protection, participants emphasized that overregulation must not become a barrier to growth. A forthcoming legislative effort to replace legacy digital laws was noted as a potential opportunity to embed principles of agile, consultative AI governance.There was emphasis that any regulation of AI must be light-touch, adaptive, and innovation-compatible. The focus, attendees agreed, should be on enabling trusted innovation rather than building constraints for an evolving technology.
AI and Intellectual Property: Rethinking Creative Incentives in the Algorithmic Age
The discussion further highlighted the need for legal frameworks that can account for AI-assisted creativity, shared value chains, and collaborative data ecosystems. Questions around training data legality, regurgitation of copyrighted content, and licensing models for large-scale AI training were examined.Participants also called for greater transparency in AI model architecture and training processes—not to constrain developers, but to build trust and shared accountability between rights-holders, innovators, and users.
Data and Competition: Building India’s Competitive Edge in the AI Economy
In discussions around data access and competition, experts noted that AI’s success increasingly hinges on access to high-quality, diverse, and multilingual datasets, as well as affordable compute. There was strong support for bolstering and building upon open public data platforms released by the Government, and frameworks that facilitate voluntary data sharing between public and private actors. Participants cautioned against replicating ex-ante regulatory models from other jurisdictions, which have shown limited success in spurring innovation.
Seizing the AI Opportunity: An India-Centric Model of Growth
The event titled, “Ingest, Generate, Compete: India’s Roadmap for AI, IP, and Competition,”brought together key stakeholders from government, industry, academia, and civil society to explore India’s potential to lead in AI development while navigating legal and economic frameworks for growth.Held under Chatham House rules, the convening focused on how India can seize the AI opportunity—by enabling innovation at scale, securing equitable access to data and compute, and establishing legal clarity without stifling experimentation.
The Esya Centre and the AI Knowledge Consortiumreaffirmed their commitment to fostering an evidence-based dialogue on AI governance—one that foregrounds innovation, equity, and Indian technological leadership.
