The Algorithmic Watch
THE Supreme Court of Zimbabwe’s rejection of AI-generated submissions by seasoned lawyer Professor Welshman Ncube may have set a legal precedent, signalling that while artificial intelligence can aid legal work, it cannot replace human accountability, accuracy, and ethical responsibility in court proceedings.
On July 5, Ncube issued an apology to the Supreme Court of Zimbabwe after a legal brief submitted under his direction was found to contain fabricated case law and misinterpretations generated by artificial intelligence.
The error occurred in the case Pulserate Investments (Pvt) Ltd v Andrew Zuze and Others [SC202/25], where Ncube was representing the appellant.
The court filing listed twelve cases that were either non-existent or irrelevant to the legal arguments presented.
On October 30, as reported by The Herald, the Supreme Court declared that legal arguments filed by Professor Welshman Ncube, generated with the assistance of artificial intelligence and citing twelve fictitious judgments, were invalid and must be treated as a nullity.
Justices Susan Mavangira, Felistus Chatukuta, and Hlekani Mwayera dismissed Ncube’s position, accepting Advocate Thabani Mpofu’s submissions that the defective heads of argument held no legal weight.
Professor Ncube admitted that the errors stemmed from a graduate researcher he had tasked with sourcing case law, who relied on AI without verifying its output.
He acknowledged that he had trusted the researcher’s work without considering the possibility of such a grave error, and also apologised to Advocate Mpofu for the inconvenience caused by verifying the non-existent citations.
Legal experts said the superior court’s rejection of the submissions underscores the importance of professional diligence and the accuracy of legal arguments presented before the bench.
AI Cases Worldwide
United Kingdom
In the United Kingdom, the High Court urged senior lawyers to act swiftly to curb the misuse of artificial intelligence after multiple fake case citations emerged.
In a ruling, Dame Victoria Sharp warned that AI misuse threatens justice and could result in sanctions, including contempt of court or referral to the police.
South Africa
South African courts have also warned lawyers that submitting AI-generated fictitious citations will trigger mandatory referral to the Legal Practice Council.
In Northbound Processing (Pty) Ltd v The South African Diamond and Precious Metals Regulator (Case No. 2025-072038), the Gauteng High Court cautioned that neither good intentions nor genuine apologies would excuse the fundamental breach of professional duty when lawyers present non-existent cases to court.
United States
According to Reuters, U.S. personal injury law firm Morgan & Morgan sent an urgent notice to its more than 1,000 lawyers in February, warning them that artificial intelligence could invent fake case law and that using fabricated information in court filings could lead to dismissal.
A federal judge in Wyoming also threatened to sanction two lawyers at the firm who included fictitious case citations in a lawsuit against Walmart (WMT.N).
One of the lawyers admitted in court filings that he had used an AI program that ‘hallucinated’ the cases and apologised for what he called an inadvertent mistake. – IOW Data.
