How can courts and tribunals manage AI litigants?
Photo by Levart_Photographer on Unsplash.
Professor Genevieve Grant, Director of the Warren Centre for Civil Justice at Monash University’s Faculty of Law, was featured in an ABC News story by reporter Jade Toomey examining the risks and responsibilities of self-represented litigants using generative artificial intelligence (AI) in legal proceedings.
The article follows a notable decision by Australia’s Fair Work Commission (FWC), which issued a costs order against a self-represented applicant who relied on AI-generated legal advice.
The pitfalls of prompts by non-lawyers
Commenting on the reliance of unrepresented litigants on tools like ChatGPT and Claude, Professor Genevieve Grant highlighted critical limitations in how non-lawyers interact with generative AI.
"If those prompts are inexpert, then the results will be as well," Professor Grant explained, drawing attention to what researchers describe as the "sycophantic tendency" of AI models.
This behavioural trait leads AI chatbots to agree with and reinforce a user's initial assumptions or arguments, regardless of whether they hold legal merit.
For self-represented litigants seeking validation, this feedback loop can create false confidence, encouraging individuals to pursue meritless claims through courts and tribunals.
Background: the Fair Work Commission ruling
The ABC report centred on a case involving a former ALDI employee, Sadnan Khan, who took his unfair dismissal claim to the Fair Work Commission while representing himself.
Mr. Khan used AI tools as a "legal advisor" to analyse submissions and draft filings.
FWC Deputy President Michael Easton ordered Mr. Khan to pay $1,230 toward ALDI’s legal fees, citing unreasonable conduct in pursuing a case built on flawed AI output.
Deputy President Easton noted that had the applicant critically read his own AI-generated submissions, he would have recognised his claim lacked substance.
In response to a 40 per cent surge in cases driven in part by generative AI adoption, the Fair Work Commission announced new regulatory requirements.
Effective 20 October, applicants filing before the FWC will be required to:
- Formally disclose any use of generative AI in preparing filings.
- Independently verify all facts, legal authorities, hyperlinks, and witness evidence cited in their submissions.
Implications for civil justice and legal technology
The decision highlights a central focus of the Warren Centre for Civil Justice to examine the evolving intersection of legal technology, empirical justice research, and court operations.
While generative AI tools offer the potential to bridge access-to-justice gaps for individuals who cannot afford legal representation, this case highlights that AI is a tool and not a substitute for legal judgment.
Without critical oversight, unguided AI reliance poses financial risks for litigants and strains on administrative tribunals.
Read the full report by Jade Toomey on