Under Section 2(d) of the Indian Copyright Act, 1957, an "author" is explicitly defined in relation to various works (literary, dramatic, musical, or artistic). Historically, the law has always presumed a human creator.

When a lawyer or a machine learning system uses an LLM (Large Language Model) to generate a unique summary of an Indian Bare Act or a complex Supreme Court judgment, where does the copyright sit?

The AI Tool? Current Indian legal frameworks do not recognize non-human entities as authors capable of holding intellectual property rights.

The Prompt Engineer/User? If a user provides highly specific, complex, and creative prompts to guide the AI, do they cross the threshold of "originality"?

The Public Domain? Does the output instantly belong to everyone?

⚖️ The Threshold of Originality: Indian jurisprudence (notably cases like Eastern Book Company v. D.B. Modak) establishes that a "modicum of creativity" and a "skill and judgment" approach are required to claim copyright over compiled legal data. Merely clicking "generate" on a generic prompt rarely meets this standard.

3 Critical IP Risks for Legal Tech Platforms
If you are building, managing, or contributing to legal knowledge databases and AI-assisted tools, three major challenges stand out:

1. Data Scraping & Fair Dealings
Training AI models requires feeding them millions of data points—including copyrighted legal commentaries and private case analyses. In India, Section 52 of the Copyright Act governs "Fair Dealing." Whether scraping legal texts for AI training falls under research or commercial infringement remains a high-stakes battlefield.

2. Derivative Works vs. Original Summaries
Compiling Bare Acts is public domain work. However, creating a highly structured, analytical dashboard or a cross-referenced index introduces a layer of formatting and structure. If an AI duplicates that precise organizational structure, it risks infringing on the compiler's derivative copyright.

3. The "Hallucination" Liability
From an ethical standpoint, if an AI tool creates a brilliantly structured blog post or brief that cites a hallucinated (fictional) case law, the liability for copyright misrepresentation and professional negligence falls squarely back on the human publisher.