WhatsApp Channel Join Now
Telegram Group Join Now

Relevance: GS-II (Gov Policies); GS-III (Science & Tech, IT, AI, Intellectual Property Rights) Source: Delhi High Court Rulings / Tech Updates, 2026

1 · What is the news?

A major Indian news agency (ANI) recently sued OpenAI (the makers of ChatGPT), claiming, “You are stealing our copyrighted news articles to train your AI!”
However, the Delhi High Court delivered a landmark ruling. The judge stated that when an AI reads and learns from publicly available internet content, it is doing it as an “internal process.” This legally qualifies as “fair dealing” for research. For millions of hardworking Indian students fighting a broken education system, this is a massive victory. Tools like ChatGPT have become free, accessible equalizers for studying. If the court had banned this, AI would become locked behind expensive paywalls, hurting the youth’s right to accessible knowledge.

2 · How Did the AI Win the Case?

Step 1: The AI Training Process
AI models (like ChatGPT) must “read” millions of webpages and articles to learn human language, facts, and grammar.
▼
Step 2: The Copyright Lawsuit
ANI filed a case in the Delhi High Court, arguing that OpenAI was copying its hard work without permission or payment.
▼
Step 3: The ‘Internal Process’ Defense
The Court noted that AI training is an internal process. The AI does not republish the exact articles to users; it just learns the underlying patterns.
▼
Step 4: The ‘Fair Dealing’ Victory
The judge ruled that this qualifies as “fair dealing” for research, protecting AI developers and ensuring free tech access for Indian youth!

3 · Key Legal & Tech Concepts Explained Simply

Section 52(1)(a)
The ‘Fair Dealing’ Rule
Indian Copyright law says using copyrighted work for private research is NOT a crime. The court beautifully expanded “research” to include machine learning by an AI!
LLMs Explained
Large Language Models
These are foundation models (like ChatGPT) trained on huge amounts of internet text. They don’t just “copy-paste”; they mathematically process data to understand language.
Idea-Expression Dichotomy
Facts vs. Exact Words
Copyright law protects the exact words (expression) an author uses, but it NEVER protects the underlying facts or ideas. If AI learns a fact from an article, it isn’t stealing!
Democratizing Education
Protecting the Youth
Amidst paper leaks and a broken system, AI acts as a free tutor. Restricting AI training would erect expensive paywalls, denying poor students their right to cutting-edge learning.

UPSC Prelims Quick Facts: Copyright Laws
The Copyright Act, 1957 Section 51 defines what constitutes an infringement (stealing), while Section 52 lists out the legal exceptions like “fair dealing” for research, criticism, or review.
Updating Construction A legal doctrine where courts interpret old laws (like a 1957 act) to fit modern technologies. Here, “research” was updated to include “machine learning.”
OpenAI Setup OpenAI Inc. is the parent developer, while OpenAI OpCo LLC operates as the commercial arm handling the business (and the lawsuits).
Substantial Similarity To prove copyright theft, the generated output must be “substantially similar” to the original. Because AI synthesizes data, its output rarely meets this criteria.

MCQ Practice Question
Q. With reference to Intellectual Property Rights (IPR) and Artificial Intelligence in India, consider the following statements:

  1. Section 52 of the Copyright Act, 1957 allows for the “fair dealing” of copyrighted works for the purpose of private use and research.
  2. The ‘Idea-Expression Dichotomy’ in copyright law implies that underlying facts and ideas can also be heavily copyrighted along with their expressions.
  3. Large Language Models (LLMs) like ChatGPT are foundation models trained exclusively on offline, uncopyrighted government data.

Which of the statements given above is/are correct?
(a) 1 only    (b) 1 and 2 only    (c) 2 and 3 only    (d) 1, 2 and 3

Answer: (a) 1 only

  • Statement 1 — Correct: Section 52(1)(a) explicitly protects “fair dealing” (like research, review, or criticism) from being treated as a copyright infringement.
  • Statement 2 — Incorrect (the trap): Do not fall for this! Copyright only protects the expression of ideas (the exact wording). It NEVER protects the underlying facts or ideas themselves.
  • Statement 3 — Incorrect: LLMs are trained on vast amounts of publicly available internet text, much of which is indeed copyrighted, leading to the current legal debates over fair use.
Ajmal IAS APSC Foundation Course

Recent Posts

Ajmal IAS APSC Foundation BATCH
Ajmal IAS APSC Foundation Course

Start Yours at Ajmal IAS – with Mentorship StrategyDisciplineClarityResults that Drives Success

Your dream deserves this moment — begin it here.