The Supreme Court on Tuesday proposed framing guidelines for the bar and bench to regulate the use of artificial intelligence (AI) in drafting judgments and filing petitions, aiming to protect the integrity of the judicial system.
A bench of justice PS Narasimha and justice Alok Aradhe said, “We are not suggesting people not to use AI. But we should have control over the information. Ultimately it is data that is filed before the court that will be used in the judgments. We want some responsibility to be fixed in this regard.”
The observations follow the discovery of two orders—one from a trial court in Andhra Pradesh and another from the National Company Law Tribunal (NCLT) in Mumbai—that contained “fake” or “hallucinated” case citations.
“This issue impacts the integrity of the proceedings in the institution on which people repose trust,” the bench said. “If a litigant or lawyer files a judgment in the court, you cannot run away from the responsibility of supplying an unverified judgment. As regards Ethics in AI, the E-Committee of the Supreme Court is devising some guidelines.”
The court directed the Bar Council of India (BCI) to assist in the task. It also sought suggestions from attorney general R Venkataramani and senior advocate Shyam Divan, who is assisting the court as amicus curiae.
{{/usCountry}}The court directed the Bar Council of India (BCI) to assist in the task. It also sought suggestions from attorney general R Venkataramani and senior advocate Shyam Divan, who is assisting the court as amicus curiae.
{{/usCountry}}“We want BCI to constitute a committee of independent experts associated with this field and file a report before us,” the bench said.
Attorney general Venkataramani said he is framing suggestions, while Divan noted that examining the experience of other countries could help formulate guidelines. The matter will be heard later this month.
The court had previously asked Divan to assist the bench on February 27, after an Andhra Pradesh trial court order cited “non-existent” or “fake” judgments. The Andhra Pradesh high court realised in January that the citations were AI-generated and, after recording a word of caution, proceeded to decide the land dispute on its merits.
In its February 27 order, the top court stated, “We take cognizance of the trial court deploying AI-generated non-existing, fake or synthetic alleged judgments and seek to examine its consequences and accountability as it has a direct bearing on integrity of adjudicatory process.”
“At the outset, we must declare that a decision based on such non-existent and fake alleged judgments is not an error in the decision making. It would be a misconduct and legal consequences shall follow,” the order added.
On Tuesday, the same bench encountered a similar issue in an appeal against an NCLT order regarding the corporate insolvency resolution process (CIRP)—a recovery mechanism for creditors—involving Essel Infraprojects.
Senior advocate Madhavi Divan, representing suspended company director Pooja Ramesh Singh, told the bench that the NCLT’s Mumbai order relied on “hallucinated” Supreme Court judgments.