Legal professionals should consider public expectations for their role, says Joshua Yuvaraj
Dr. Joshua Yuvaraj has hypothesised that any efficiency increase due to artificial intelligence (AI) will meet a correspondingly greater verification cost, which renders the value of such tools negligible for the tasks they automate, including research, drafting, and document review.
“This is because the more we trust AI, the more costly mistakes are to clients, and the more important it becomes to verify the outputs,” he said in an article published on Newsroom.
Yuvaraj, a senior lecturer at the University of Auckland Faculty of Law and co-director of the New Zealand Centre for Intellectual Property, discussed this hypothesis in his paper for the Monash University Law Review, titled “The Verification-Value Paradox: A Normative Critique of Gen AI in Legal Practice.”
According to Yuvaraj, while AI providers advertise their tools as streamlining legal practice and making legal advice more affordable and efficient for clients, generative AI and possibly other AI tools rely on two structural flaws.
First, he stated that AI models do not understand whether facts are accurate and thus make mistakes. He referred to a study finding that AI tools designed for law firms ‘hallucinated’ by inventing cases and providing other false information between 17 and 33 percent of the time.
Second, Yuvaraj said AI tools frequently lack transparency and operate as ‘black boxes’ preventing their users from learning how the tools have arrived at their outcomes.
In the article, Yuvaraj stressed that lawyers should uphold professional values such as integrity and stand by the accuracy of the information they provide, including by meticulously verifying what an AI model generates.
He noted that judges worldwide have disapproved of lawyers submitting inaccurate AI-produced material, including fabricated cases and misquotes of genuine cases, and have highlighted the need to ensure the accuracy of all content provided to courts.
Yuvaraj said a UK judge raised the possibility of holding lawyers criminally liable for giving the courts false AI-generated information.
He added that lawyers might face negligence lawsuits based on inaccurate AI-produced advice. He noted that Deloitte had to partly refund the Australian government for providing a report that might have contained AI-generated mistakes.
Yuvaraj acknowledged the value of this technology in certain situations. For lawyers insistent on utilising AI, he made two suggestions in the article.
First, he urged lawyers to think twice about the risk of reputational damage before resorting to AI.
“[I]magine a client’s embarrassment if, having paid top dollar to a firm to provide bespoke advice/defend you in complex criminal proceedings, they find out their lawyer has a) used AI without telling them, and/or b) not vetted the content so it produces mistakes that could cost them millions, or even prison time,” Yuvaraj said in the article.
Second, he asked lawyers to consider public expectations for the role of legal professionals in an increasingly uncertain world. According to him, people expect lawyers to prioritise the truth and act as servants, rather than acting in a self-serving manner.
“Law is about serving others first – which cuts against the grain of shortcut-taking that has caught so many lawyers around the world using AI in court proceedings,” Yuvaraj said.
