Justice Gavai Flags AI Risks: When ChatGPT Gets Legal Facts Wrong

Senior Research Associate, Dona Mathew was quoted in a Medianama article looking at the ethical concerns about the use of generative AI within the Indian judicial system flagged by Supreme Court judge Justice BR Gavai.

Dona pointed out two similar cases in Indian courtrooms where artificial intelligence played spoilsport and perverted the natural course of justice.

In the Punjab and Haryana High Court, some judge had said that he had been using ChatGPT to figure out bail jurisprudence. Even in January 2025, some judge in a trial court in Karnataka ended up using ChatGPT and cited a wrong precedent. And then both the petitioner and defendant said that they had not actually cited those cases in their original arguments.

Read the full story here.