In courtrooms across Australia, an invisible force is quietly disrupting centuries of legal tradition. Artificial intelligence, once confined to science fiction, has made a stealthy entrance into the halls of justice. Lawyers, enticed by the promise of enhanced efficiency, have begun to embrace AI-powered tools to draft documents, analyze cases, and even generate evidence. But as AI’s influence grows, judges are confronting an unsettling reality: the potential for fake cases, erroneous legal citations, and affidavits of dubious origin.
The extent of AI’s penetration into the legal realm remains shrouded in uncertainty. A Thomson Reuters survey of nearly 900 Australian legal professionals hinted at a profession grappling with change. While 40% worked at firms experimenting with AI, caution tempered their approach. Only 9% actively relied on AI in their daily work. Yet a third expressed a desire for an AI legal assistant, signaling a growing appetite for algorithmic aid.
The Specter of Hallucinated Citations
As AI-powered tools like ChatGPT gain traction among lawyers, troubling incidents have emerged. In one case, a Melbourne solicitor relied on Leap, a legal software with generative AI capabilities, to prepare a list of prior cases requested by the court. To the bafflement of the judge and associates, the cited cases proved untraceable. The lawyer later confessed that Leap had “hallucinated” the citations, conjuring precedents out of digital thin air.
We encourage the correct and ethical use of our integrated AI products, and have implemented a range of mitigation, education and professional development measures.
– Christian Beck, CEO of Leap
Affidavits Under Scrutiny
Courts are also casting a wary eye on affidavits and personal references. In a criminal case, the Australian Capital Territory Supreme Court questioned the authenticity of an offender’s character reference, noting signs it was authored by an AI system. The telltale clue? A phrase asserting the writer had known the offender “both personally and professionally for an extended period”—an claim the court found doubtful coming from the offender’s own brother.
Judicial Wariness and New Limits
Faced with the specter of AI-tainted evidence, judges are erecting guardrails. Justice Kristen Walker of the Victorian Court of Appeal, in a university dismissal case, opted not to reproduce suspect case citations, lest they “contribute to the problem of [AI] inventing case citations.” The NSW Supreme Court has gone further, issuing a practice note barring the use of generative AI for affidavits, witness statements, and other evidentiary material.
Train people in where [AI] is useful, because once you start using it in a conscious way, you realise it’s actually not good at these sorts of things.
– Professor Jeannie Paterson, University of Melbourne
The Path Forward: AI Literacy and Ethical Guardrails
As the legal world grapples with AI’s ascendance, experts stress the importance of AI literacy for lawyers. Professor Jeannie Paterson of the University of Melbourne sees the recent missteps as a wake-up call. “I think this is an AI literacy problem, as much as a slack solicitor problem,” she remarks. Paterson advocates for training lawyers to understand AI’s strengths and limitations, lest the legal system “implode” under the weight of algorithmic illiteracy.
Legal oversight bodies, from the Victorian Legal Services Board to the Queensland Legal Services Commissioner, are also sounding the alarm. They emphasize the duty of lawyers to provide accurate information and exercise sound judgment—qualities that AI, for all its computational prowess, cannot replicate. As Megan Mahon, the Queensland Commissioner, puts it: “It is critically important that anyone seeking legal advice ensures they are engaging with a qualified and licensed legal practitioner.”
The path forward, it seems, lies in striking a delicate balance. Lawyers must learn to harness AI’s potential while remaining vigilant against its pitfalls. Courts and regulators have a vital role to play in setting ethical guidelines and enforcing standards. The stakes could not be higher: the integrity of the justice system hangs in the balance.
As the legal world stands at this crossroads, one thing is certain: the age of AI in law has dawned, and there can be no turning back. The challenge now is to navigate this uncharted territory with wisdom, ethics, and an unwavering commitment to the truth. For in the quest for justice, there can be no room for hallucinated precedents or algorithmic sleight of hand. The future of the law, and the faith of those it serves, will be forged in the crucible of this technological revolution.
