Citing AI-Generated Fake Precedents Is Advocate Misconduct; Judgments Based On Them Are Void: Supreme Court
📋 Table of Contents
- 1. Introduction: A Wake-Up Call for the Legal Profession
- 2. What Exactly Happened? The Case That Shook the Legal World
- 3. The Supreme Court's Landmark Ruling: Zero Tolerance for Fake AI Precedents
- 4. Why Citing AI-Generated Fake Judgments Is Professional Misconduct
- 5. Why Judgments Based on Fake Precedents Are "Void"
- 6. The Role of the Bench: Judges Cannot Escape Scrutiny Either
- 7. Bar Council of India Directed to Frame AI Guidelines
- 8. This Was Not the First Time: A Growing Menace in Indian Courts
- 9. The Global Context: AI Hallucinations in Courts Worldwide
- 10. Why AI-Generated Fake Precedents Are Dangerous for Justice
- 11. How Lawyers Can Protect Themselves and Their Clients
- 12. What This Means for Ordinary Citizens
- 13. The Future of AI in Indian Legal Practice
- 14. Conclusion: Technology Must Serve Justice, Not Subvert It
1. Introduction: A Wake-Up Call for the Legal Profession
Imagine walking into a courtroom with a brief full of case laws that sound impressive, well-reasoned, and perfectly suited to your argument. The judge reads them, relies on them, and delivers a verdict in your favor. But here is the terrifying truth — those judgments never existed. They were invented by an artificial intelligence tool. They are complete fabrications. They are what the tech world calls "AI hallucinations."
This is not science fiction. This is happening right now in Indian courtrooms. And on July 2, 2026, the Supreme Court of India drew a line in the sand. In a landmark ruling, the apex court declared that citing AI-generated fake precedents is professional misconduct on the part of advocates. Even more shockingly, the Court held that any judgment based on such fake precedents is void — it is "no decision in the eyes of the law."
This ruling is not just a warning. It is a thunderclap that will echo through every law firm, every courtroom, and every legal research desk in India. It changes how lawyers can use AI tools. It changes how judges must verify citations. And it changes what ordinary citizens can expect from the justice system. Let us understand this ruling in depth, in simple words, because this affects every single person who believes in the rule of law.
📖 Related Reading: Before we dive deeper, if you want to understand the broader framework of arrest rights and how the law protects citizens from arbitrary state action, read our detailed guide on Section 47 of BNSS – Person Arrested to Be Informed of Grounds of Arrest and Right to Bail. Understanding your fundamental rights is the first step toward justice.
2. What Exactly Happened? The Case That Shook the Legal World
The Supreme Court's ruling came in an appeal arising from insolvency proceedings. Jammu and Kashmir Bank Ltd. had filed an insolvency application under Section 7 of the Insolvency and Bankruptcy Code (IBC) against Essel Infraprojects Ltd. (EIL), which had executed a corporate guarantee in respect of credit facilities extended to Pan India Utilities Distribution Company Ltd.
The National Company Law Tribunal (NCLT), Mumbai, admitted the insolvency application on August 28, 2024, recording a default of Rs. 87.43 crore. The National Company Law Appellate Tribunal (NCLAT) subsequently affirmed the admission order on September 11, 2025. Everything seemed normal. But when the matter reached the Supreme Court, something extraordinary was discovered.
During its independent scrutiny of the case, the Supreme Court found that several judicial citations relied upon by the NCLT were either entirely fictitious or contained fabricated extracts. These were not minor errors. These were complete inventions — judgments that had never been delivered by any court, attributed to real or imaginary judges, with elaborate reasoning that sounded entirely plausible but was entirely fake.
The Supreme Court was hearing an appeal against the NCLAT's order. But instead of merely deciding the insolvency issue, the Bench comprising Justice P.S. Narasimha and Justice Alok Aradhe confronted a much larger problem — the integrity of the judicial process itself was under attack by artificial intelligence.
3. The Supreme Court's Landmark Ruling: Zero Tolerance for Fake AI Precedents
The Supreme Court did not mince words. It adopted a "zero-tolerance" approach towards AI-generated, fake, or hallucinated judicial precedents. The Court made several declarations that will reshape legal practice in India:
- It is professional misconduct for advocates to cite such judgments without verification. The Court stressed that lawyers have a professional obligation to ensure every legal authority placed before a court is genuine. Simply copying citations from an AI tool without checking them against official legal databases is not negligence — it is misconduct.
- It is a serious lapse on the part of judges to rely on such non-existent precedents while deciding cases. The Court did not spare the judiciary either. It held that judges who rely on fake or hallucinated AI-generated material as precedents commit a serious lapse that undermines the sanctity of adjudication.
- A decision based on such fake precedents is "no decision in the eyes of the law." This is perhaps the most powerful declaration. The Court said that any decision tainted by fabricated or hallucinated precedents is void — it carries no legal weight, regardless of whether the fake material directly influenced the outcome.
- "An iota of fake or hallucinated material" is enough to set aside a decision. The Court made it clear that even a trace of fabricated content undermines the entire adjudicatory process. Such decisions deserve to be set aside irrespective of whether the fake material had a direct or indirect bearing on the outcome.
📜 Key Quote from the Supreme Court:
"It is necessary for Courts to adopt a zero-tolerance mode for producing, citing or using AI-generated precedents without verification. It is a misconduct on the part of an advocate to cite such judgments without verification."
"Equally, it is a serious lapse if a judge relies on such a fake or hallucinated AI-generated material as precedents in support of the determination."
"We have no hesitation in declaring that such a decision is no decision in the eyes of the law, irrespective of whether such material had a direct or indirect bearing on the decision-making."
The Court also clarified an important point: "Our judgment shall have no bearing on the rightful use of AI, but on the presentation or reliance on fake or hallucinated material as if it were a court precedent." This means AI itself is not banned. Lawyers can still use AI for research, drafting, and analysis. But they must verify every citation. They must treat AI as a starting point, not an authority.
4. Why Citing AI-Generated Fake Judgments Is Professional Misconduct
To understand why the Supreme Court called this "misconduct" and not merely "error," we need to understand what professional misconduct means in the legal profession. Under the Advocates Act, 1961 and the Bar Council of India Rules, an advocate is an officer of the court. This is not just a title. It is a sacred trust.
When a lawyer stands before a judge and cites a case, they are making an implicit promise: "This judgment exists. It was delivered by a real court. The quotes I am reading are accurate. The legal principles I am relying on were actually laid down by a real judge." This promise is the foundation of the adversarial system. Judges rely on advocates to be honest about their citations. They do not have time to verify every single case cited in every single brief. They trust the Bar.
When an advocate cites an AI-generated fake judgment, they break this trust. They deceive the court. They waste judicial time. They potentially influence the outcome of a case based on fiction. This is not a mistake like citing the wrong page number or misquoting a line. This is presenting fabricated evidence to a court of law.
The Supreme Court's ruling makes it clear that this behavior falls squarely within the definition of professional misconduct. The consequences can include:
- Disciplinary action by the Bar Council of India — suspension or even removal from the rolls of advocates
- Contempt of court proceedings — for misleading the court
- Criminal prosecution — in serious cases, for fraud or forgery
- Professional ostracism — loss of reputation, clients, and career
- Personal liability — for damages caused to clients who lost cases because of fake citations
⚠️ The Critical Distinction: "Reasons" vs "Grounds"
The Supreme Court has drawn a crucial distinction between "reasons" for arrest and "grounds" of arrest. Similarly, in the context of AI citations, there is a distinction between using AI to generate ideas (reasons) and presenting AI-generated fake judgments as real authorities (grounds). The former may be acceptable with caution. The latter is outright misconduct. Just as Article 22(1) requires communication of actual grounds, not vague reasons, the Court now requires actual verified judgments, not AI hallucinations presented as real.
5. Why Judgments Based on Fake Precedents Are "Void"
This part of the ruling is perhaps the most legally significant. The Supreme Court declared that a decision based on fake or hallucinated precedents is "no decision in the eyes of the law." This is not just strong language. It has profound legal consequences.
In legal terms, when a court says a decision is "void," it means the decision is invalid from the very beginning (void ab initio). It is as if the decision was never made. It creates no legal rights. It imposes no legal obligations. It cannot be enforced. It cannot be relied upon in future cases. It is erased from the legal record.
The Supreme Court explained why this must be so. The integrity of the adjudicatory process is sacred. When a court delivers a judgment, it is not just deciding a dispute between two parties. It is speaking the law. It is creating precedent that other courts will follow. It is building the edifice of justice brick by brick. If even one brick is fake — if even one precedent cited in a judgment never existed — the entire structure is compromised.
The Court said: "Such decisions are to be set aside even if an iota of fake or hallucinated material enters the decision-making process, as it would violate the sanctity of adjudication." This means the test is not whether the fake precedent actually changed the outcome. The test is whether fake material entered the process at all. If it did, the decision is tainted. It must be quashed.
In the Essel Infraprojects case, the Supreme Court did exactly what it preached. It quashed the NCLT's order and restored the Section 7 insolvency application for fresh consideration. It directed the tribunal to hear the matter afresh, preferably within two weeks. The months of litigation, the NCLAT's affirmation, the parties' expenses — all of it was undone because the original decision was built on fake foundations.
6. The Role of the Bench: Judges Cannot Escape Scrutiny Either
The Supreme Court's ruling was remarkable not just for what it said about advocates, but for what it said about judges. The Court made it clear that judges who rely on fake AI-generated precedents are guilty of a "serious lapse."
This is a bold and necessary statement. In the Essel Infraprojects case, the fake citations were not caught by the NCLT. They were not caught by the NCLAT either. They were only discovered when the matter reached the Supreme Court. The apex court did not hesitate to call this out.
The Bench pointedly asked: "What about the Appellate Tribunal? The fake, non-existent judgments escaped scrutiny by the first statutory appellate tribunal. Today's courts and tribunals implicitly trust lawyers when referring to precedents cited before them."
This observation reveals a deep institutional problem. Courts and tribunals operate under enormous pressure. They handle thousands of cases. They cannot possibly verify every citation in every brief. They rely on the integrity of the Bar. When that integrity is compromised by AI-generated fake judgments, the entire system breaks down.
The Supreme Court's message to judges is clear: You cannot blindly trust citations. You must exercise due diligence. You must verify critical precedents, especially when they seem unusual or when they form the basis of your reasoning. This adds to the judicial workload, but the Court says it is necessary to preserve the sanctity of the process.
7. Bar Council of India Directed to Frame AI Guidelines
The Supreme Court did not stop at declaring the law. It also took proactive steps to prevent future misuse. The Court directed the Bar Council of India (BCI) to constitute a committee to examine the growing use of artificial intelligence in legal practice and frame comprehensive guidelines governing its use.
The committee has been tasked with several important responsibilities:
- Examining the growing use of AI in legal practice and adjudication — understanding how lawyers and judges are currently using AI tools
- Framing comprehensive guidelines — setting clear rules for when and how AI can be used in legal research, drafting, and court filings
- Recommending disciplinary measures — proposing specific penalties for lawyers who submit fake or fabricated legal authorities before courts
- Educating the legal profession — ensuring that every advocate understands the risks of AI hallucinations and the duty to verify
This is a significant development because until now, India had no specific regulations governing the use of AI by lawyers. The Advocates Act, 1961 and the BCI Rules were written long before ChatGPT existed. They do not mention artificial intelligence. The Supreme Court's direction fills this regulatory gap and ensures that the legal profession will have clear, enforceable standards for AI use.
📖 Related Reading: The Supreme Court's emphasis on proper procedure and verification mirrors the safeguards built into our criminal justice system. Learn about how Section 35 of BNSS — Arrest Rules and Notice Before Arrest ensures that police cannot arrest without following strict procedural requirements. The law always demands verification and accountability.
8. This Was Not the First Time: A Growing Menace in Indian Courts
The Essel Infraprojects case was shocking, but it was not an isolated incident. The Supreme Court itself had flagged this problem earlier. In February 2026, a Bench headed by Chief Justice Surya Kant expressed serious concern over a growing trend of lawyers filing petitions drafted with AI tools that contained non-existent judgments.
In one particularly bizarre instance, a petition cited a case called "Mercy vs. Mankind." This case never existed. It was a complete invention by an AI tool. Yet it appeared in a formal court filing. The Chief Justice called this trend "alarming."
In another case before Justice Dipankar Datta's court, multiple fabricated judgments were cited. All proved to be complete inventions. The Court had to spend valuable judicial time investigating whether these judgments existed, only to discover that they were AI hallucinations.
Earlier, in March 2026, the Supreme Court had taken cognizance of a trial court relying on alleged non-existing verdicts generated with AI. The Court said: "We take cognisance 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 the integrity of the adjudicatory process."
In the Bombay High Court, a case called Heart & Soul Entertainment Ltd. vs. Deepak (March 2026) exposed AI-generated citations. The High Court discovered that the citations were of the nature of "repetitive content" and "formatting patterns" — telltale signs of AI-generated content. The High Court imposed a penalty of Rs. 50,000 and called such actions a "hurdle in the swift delivery of justice." The Supreme Court upheld this penalty, calling AI misuse a "menace" to the legal profession.
In Gummadi Usha Rani & Anr. v. Sure Mallikarjuna Rao & Anr., the Supreme Court found that a trial court had relied on four non-existent Supreme Court judgments:
- Subramani v. M. Natarajan (2013) 14 SCC 95 — Fake
- Chidambaram Pillai v. SAL Ramasamy (1971) 2 SCC 68 — Fake
- Lakshmi Devi v. K. Prabha (2006) 5 SCC 551 — Fake
- Gajanan v. Ramdas (2015) 6 SCC 223 — Fake
None of these cases exist in any official legal database. They were complete fabrications.
In Punam Pandey vs. Former SHO, PS: Mehrauli & Ors., a Delhi Court rebuked a complaint application as having "more technical intervention and less of human mind contribution." The Court observed that the pleadings "certainly do not make any sense and fail to convey anything else" and imposed a fine of Rs. 20,000 for wasting judicial time.
9. The Global Context: AI Hallucinations in Courts Worldwide
India is not alone in facing this problem. The misuse of AI-generated fake citations has become a global crisis in the legal profession. The most famous case is Mata v. Avianca, Inc. from the United States in 2023. Two lawyers used ChatGPT to draft a legal brief and submitted it to a federal court. The brief contained entirely fabricated case citations. The judge described the pleadings as "gibberish" and the lawyers as having acted in "bad faith." They were fined $5,000 and referred to the bar association for disciplinary action.
In the United Kingdom, the case of Jonathan David Sheppard vs. Jillion LLC reached the Qatar Financial Centre Civil and Commercial Court. The Court found that using AI without thorough double-checking "ordinarily amounts to reckless conduct" and in some cases "amounted to intentional conduct."
In Massachusetts, a lawyer was sanctioned $2,000 for citing fictitious cases produced by an AI tool. The judge wrote: "Many harms flow from the submission of fake opinions."
In California, two law firms were fined $31,000 for submitting a brief with fake citations generated by AI. The judge criticized the firms for undisclosed AI use, saying it misled the court.
In Utah, a lawyer named Richard Bednar was sanctioned by the Court of Appeals for submitting a brief with fake ChatGPT citations, including a non-existent case called "Royer v. Nelson." He was ordered to pay attorney fees, refund client fees, and donate $1,000 to a legal non-profit.
These cases show that the problem is universal. AI tools like ChatGPT are trained to generate plausible-sounding text. They are not trained to verify facts. When asked for legal precedents, they often invent cases with real-sounding names, real-sounding judges, and real-sounding reasoning. To a busy lawyer, these can look legitimate. To a busy judge, they can look authoritative. But they are fiction. And the consequences of relying on fiction in a court of law are devastating.
10. Why AI-Generated Fake Precedents Are Dangerous for Justice
To truly understand why the Supreme Court took such a strong stand, we need to understand what is at stake. AI-generated fake precedents are not just embarrassing mistakes. They are a direct threat to the foundations of justice.
They undermine the doctrine of stare decisis. Stare decisis is the principle that courts must follow precedents laid down by higher courts. It ensures consistency, predictability, and fairness in the law. When fake precedents enter the system, they pollute this doctrine. A lower court might follow a fake precedent, creating a chain of bad decisions. Future lawyers might cite the fake precedent, believing it to be real. The entire edifice of case law becomes contaminated.
They waste enormous judicial resources. When a judge discovers that a cited case does not exist, they must stop everything and investigate. They must check legal databases. They must ask the parties for clarification. They may need to adjourn the hearing. In a system already burdened by millions of pending cases, this is a luxury India cannot afford.
They can determine the outcome of real cases. In the Essel Infraprojects case, the NCLT's admission of the insolvency application affected the rights of creditors, employees, and stakeholders worth crores of rupees. If that decision was influenced by fake precedents, the financial and human consequences are staggering.
They erode public trust in the judiciary. The average citizen believes that court judgments are based on real law, real facts, and real reasoning. If people learn that courts are relying on AI-generated fiction, their faith in the justice system will collapse. And without public trust, no judiciary can function.
They create a moral hazard. If lawyers know they can get away with citing AI-generated fake judgments, some will be tempted to do so deliberately — to strengthen weak arguments, to confuse opponents, or to manipulate outcomes. The Supreme Court's zero-tolerance approach is designed to kill this temptation before it spreads.
📖 Related Reading: The sanctity of judicial process depends on every participant following the rules. Just as fake precedents destroy trust, unlawful arrests destroy citizen confidence in the system. Understand your rights under Section 57 of BNSS — Rights After Arrest, which mandates that every arrested person must be produced before a Magistrate without unnecessary delay. Procedure matters.
11. How Lawyers Can Protect Themselves and Their Clients
The Supreme Court's ruling does not ban AI. It bans unverified AI. Lawyers can still use AI tools, but they must follow strict protocols. Here is what every advocate should do:
- Verify every citation against official legal databases. Do not trust AI-generated case names. Check them on SCC Online, Live Law, Indian Kanoon, or the official Supreme Court website. If a case cannot be found, do not cite it.
- Read the full judgment before citing it. AI tools often generate plausible-sounding extracts that do not actually appear in the real judgment. Read the full text to confirm that the reasoning attributed to the case is actually there.
- Be suspicious of unusual case names. If a case name sounds odd — like "Mercy vs. Mankind" — be extra cautious. Verify it immediately. Real case names follow patterns. Fake ones often do not.
- Keep records of your research. Document where you found each citation. If a court questions a citation, you must be able to show that you verified it from a legitimate source.
- Train junior lawyers and interns. Many fake citations enter the system through inexperienced researchers who do not know the risks. Every law firm should have a mandatory verification policy.
- Disclose AI use to clients. Clients have a right to know how their case is being prepared. If you use AI for research, tell your client. If you discover that AI-generated content was used in a filing, correct it immediately.
- When in doubt, leave it out. If you cannot verify a citation, do not cite it. A weaker argument based on real law is better than a stronger argument based on fiction.
💡 Best Practice: Think of AI as a junior research assistant who is brilliant at finding ideas but terrible at getting facts right. You would never submit a brief based solely on a junior's work without checking it. Treat AI the same way. It is a tool, not an authority.
12. What This Means for Ordinary Citizens
You might be thinking: "I am not a lawyer. Why should I care about AI-generated fake precedents?" The answer is simple: because this affects the justice you receive.
If you are a party to a lawsuit, your lawyer's research determines the strength of your case. If your lawyer relies on fake precedents, your case could be dismissed. You could lose property, liberty, or money because of an AI hallucination.
If you are accused of a crime, the precedents cited in your bail application or defense brief could determine whether you walk free or remain in custody. Fake precedents could send an innocent person to jail or keep a guilty person free.
If you are a creditor in an insolvency case, fake precedents could affect whether your debt is recovered. The Essel Infraprojects case shows how billions of rupees in financial claims can be affected by fabricated citations.
As a citizen, you have the right to ask your lawyer: "Have you verified every citation in my case?" You have the right to demand that your lawyer uses only real, verified legal authorities. You have the right to expect that the judge deciding your case is relying on genuine precedents, not AI fiction.
The Supreme Court's ruling protects you. It ensures that the justice system remains grounded in reality, not in the imaginary outputs of a machine.
13. The Future of AI in Indian Legal Practice
The Supreme Court's ruling is not the end of AI in law. It is the beginning of regulated, responsible AI use. The Court itself acknowledged that AI can be a powerful tool when used correctly. The problem is not AI. The problem is unverified AI.
In the future, we can expect several developments:
- AI-specific regulations from the BCI — clear guidelines on disclosure, verification, and penalties
- AI-assisted legal research tools with built-in verification — tools that automatically check citations against official databases before including them in drafts
- Mandatory disclosure requirements — lawyers may be required to declare whether AI was used in preparing a brief, and if so, how citations were verified
- Judicial training on AI — judges may receive training on how to spot AI-generated content and how to verify suspicious citations
- Standardized citation formats — courts may require citations to include hyperlinks or database references that make verification easier
- AI auditing of court filings — courts may use AI tools to automatically flag potentially fake citations before they reach the judge
The Artificial Intelligence (Ethics and Accountability) Bill, 2025, which has been introduced in Parliament, may also provide a broader regulatory framework for AI use across sectors, including the legal profession.
The goal is not to ban AI. The goal is to ensure that AI serves justice, not subverts it. When used responsibly, AI can help lawyers find relevant cases faster, draft better arguments, and serve clients more efficiently. But this can only happen if every citation is verified, every extract is confirmed, and every judgment is real.
📖 Related Reading: The Supreme Court's approach to AI regulation reflects a broader judicial philosophy — that power must always be balanced by accountability. This same principle applies to search warrants under Section 96 of BNSS — Search Warrants, where the law ensures that no search happens without judicial oversight and proper safeguards. Accountability is the cornerstone of justice.
14. Conclusion: Technology Must Serve Justice, Not Subvert It
The Supreme Court's July 2026 ruling is a watershed moment for the Indian legal profession. It sends an unambiguous message: AI is welcome, but fiction is not. Lawyers who use AI must verify every citation. Judges who review briefs must exercise due diligence. And the Bar Council must provide clear guidelines to ensure that technology enhances rather than undermines the justice system.
The ruling also reminds us of a deeper truth. Justice is a human enterprise. It requires judgment, wisdom, integrity, and accountability. AI can assist, but it cannot replace the human mind. A machine that "hallucinates" case laws cannot grasp the moral weight of a decision that affects a person's liberty, property, or life. Only a human judge, guided by real precedents and real reasoning, can do that.
The Essel Infraprojects case will be remembered not for its insolvency outcome, but for the principle it established: the sanctity of adjudication is non-negotiable. Even an iota of fake material destroys the legitimacy of a judicial decision. And those who introduce fake material — whether deliberately or negligently — must face consequences.
For every lawyer reading this, the message is clear: Verify, verify, verify. For every citizen reading this, the message is equally clear: Demand verification. Ask your lawyer how they researched your case. Ask whether every citation is real. Because in the end, justice depends on truth. And truth cannot be hallucinated.
🛡️ Final Takeaway: The Supreme Court has declared zero tolerance for AI-generated fake precedents. Citing them is misconduct. Judgments based on them are void. The Bar Council must frame guidelines. And every lawyer, judge, and citizen must ensure that the law remains rooted in reality, not in the imaginary outputs of artificial intelligence.
COMMENTS