Zero Tolerance for AI Generated Case Laws Without Verification: Such decisions are no decisions

The Supreme Court in Vijay Ghanshyam Gadiya Versus Union Of India & Anr. - 2026 (9) TMI 363 - Supreme Court  has laid down a landmark principle: reliance on unverified, AI generated case laws or citations is impermissible and can itself be a ground to set aside judicial or quasi judicial orders. This ruling reflects the judiciary's growing concern over the phenomenon of "AI hallucination," where non existent precedents are fabricated and presented as authoritative.

The Supreme Court drew upon the precedent in Pooja Ramesh Singh v. Jammu and Kashmir Bank Ltd., which had earlier established a doctrine of zero tolerance for fake citations. In the judgement it is declared that "Such decisions are no decision in eyes of law irrespective of whether fake material had direct or indirect bearing on decision-making".

In the present case, the penalty order was found to have relied on dubious AI generated material. The Court clarified that even if such fabricated references did not directly influence the decision making process, their mere presence invalidates the order in the eyes of law. Consequently, both the original order and the High Court's affirmation were set aside, with directions for fresh proceedings before an officer of equal rank.

Significantly, the judgment also directed the appointing authority to initiate appropriate action against the officer responsible for the flawed order. This underscores the accountability of adjudicating officers and the need for rigorous verification before relying on AI assisted tools in judicial or quasi judicial contexts.

The ruling sends a strong signal: AI can assist, but cannot substitute human diligence. Case laws and precedents form the backbone of legal reasoning, and any compromise on their authenticity undermines the integrity of the justice system. By instituting a zero tolerance doctrine, the Supreme Court has safeguarded the sanctity of adjudication while cautioning against blind reliance on emerging technologies.