Skip to content
Submit incident
AII-1002-09829
ConfirmedA court, a regulator or the organization itself has stated the core facts, or the record cites a court or regulator source.Supporting sources are pending verification.

Judge criticizes lawyers for filing AI-generated documents with false citations in murder case

Lawyers defending a murder case in Melbourne filed court documents containing AI-generated false citations and nonexistent laws without verifying the content. A judge criticized the practice as undermining the administration of justice and said all use of AI must be independently verified.

14 August 2025 Β· Amelia Beech Β· Reliability and hallucination
14 August 2025
AII-1002-09829
Quick facts
Category
Reliability and hallucination
Country
Australia
Date
14 August 2025
Drafted by
Raidu Incident Detection

What happened

Lawyers acting for a 16-year-old boy accused of murder filed court submissions containing misleading information created by AI without proper verification. The documents included references to nonexistent case citations and inaccurate quotes from a parliamentary speech.

When the court could not locate the referenced material, the defence admitted it had used AI. The documents were not signed by barristers or solicitors when filed. Prosecutors also did not verify the information and created their own submissions based on the flawed defence documents.

After the defence apologised and re-filed documents, these revised submissions again contained made-up laws and were not reviewed by either side before submission. The defence eventually filed documents without AI inaccuracies after a second apology.

Reported impact

Justice James Elliott stated that 'the ability of the court to rely on submissions was fundamental to the administration of justice' and that use of AI without careful oversight would 'seriously undermine this court's ability to deliver justice'.

Index analysis

The AI system generated nonexistent case citations and inaccurate quotes from parliamentary speeches. Revised submissions referred to legislation that did not exist, including references to acts that were repealed that never occurred.

Governance controls

The Supreme Court of Victoria has guidelines on AI that state use of AI must not 'directly mislead another participant in the litigation process as to the nature of any work undertaken or the content produced by that program'. Justice Elliott stated 'It is not acceptable for AI to be used unless the product of that use is independently and thoroughly verified'.

Classification

Category
Reliability and hallucination
Industry
Government, security and defence
Country
Australia
Harm status
Actual harm
Intent
Unintentional
Timing
Unknown
Harm types
Public interest
Harmed
General public
Failure modes
Fabricated citation, Hallucination, Overreliance
AI systems
Unspecified large language model

Sources and evidence

Sources not yet published

This record was researched and written by the Index from original reporting. Its sources are still being verified by an editor, and they are listed here once that check is complete. Until then the record carries no reviewer.

How sources are verifiedSuggest a source for this incident