Technology and artificial intelligence legal advisory

Artificial Intelligence & Intellectual Property Rights: Who Owns Algorithmic Outputs in India?

The explosive adoption of generative artificial intelligence (GenAI) and large language models (LLMs) across software engineering, digital media, and automated research has triggered unprecedented questions within Indian intellectual property jurisprudence.

Key Legal Questions Confronting AI Deployments

Enterprises integrating AI tools must address three fundamental IP and liability vectors:

  • Authorship and Copyrightability: Under Section 2(d) of the Indian Copyright Act, 1957, copyright requires a human author. Purely autonomous outputs generated by machine algorithms lack copyright protection unless substantial human creative expression is demonstrated.
  • Copyright Infringement in Training Data: The unauthorized scraping and commercial utilization of copyrighted works to train commercial AI systems poses significant exposure under Indian copyright law, where ‘fair dealing’ exceptions are more restrictive than US ‘fair use’.
  • Trade Secret Vulnerabilities: Staff feeding proprietary source code, non-public client financial models, or internal communications into public AI chatbots risk forfeiting trade secret protections and breaching enterprise non-disclosure agreements (NDAs).

Practical Safeguards for Enterprises

To mitigate risk, companies should immediately establish an internal Enterprise AI Governance Policy, secure enterprise enterprise-grade instances guaranteeing zero data retention for training, and introduce robust indemnification clauses into software development vendor contracts.

Looking for something?

Hey, I’m Neha.

Adv. Neha Vashisth

Advocate practising across corporate, technology, data protection, intellectual property, and cross-border matters. This journal collects practical legal analysis and commentary on Indian law.