AI-Assisted Litigant Wins Australian Employment Ruling
Last week, Australia's Fair Work Commission ruled that Macquarie University computing academic Gregory Baker should be treated as an ongoing part-time employee after he represented himself with help from trained AI agents. The Conversation reports that Baker used an AI tool to identify a possible pathway for converting his casual employment status, then pursued the dispute without a lawyer.
Australia's Fair Work Commission ruled that Macquarie University computing academic Gregory Baker should be treated as an ongoing, part-time employee, after the university declined his request to convert from casual employment. Baker represented himself in the tribunal proceeding and has said he used trained AI agents to assist his case, according to an August 18 analysis in The Conversation.
Baker had taught computer science across consecutive semesters from 2023 through 2025. The Conversation reports that in November 2025 he notified Macquarie University that he considered his work no longer met the requirements for casual employment. After the university rejected that notification, he lodged a Fair Work Commission dispute without a lawyer in December 2025.
The parties did not reach agreement, and the matter proceeded to arbitration on May 12. The decision was issued in August. The Conversation notes that the Fair Work Commission is a tribunal rather than a court, but that its orders are legally binding.
AI assistance and legal representation
According to The Conversation, an AI tool first alerted Baker to the possibility that he could seek conversion from casual to permanent part-time employment. The article characterizes the result as an example of a highly capable user employing AI tools effectively, rather than evidence that AI has replaced lawyers.
That distinction is material for practitioners building or evaluating legal AI systems. Employment disputes depend on procedural rules, statutory tests, evidence, and the user's ability to verify outputs and present a coherent case. In comparable high-stakes workflows, generative systems can lower the cost of issue spotting, document analysis, and drafting, but incorrect legal citations or misunderstood procedural requirements can still create substantial risk.
The case also differs from the separate English debt-recovery matter involving Garfield AI, in which an AI law firm prepared litigation materials and retained a human barrister for trial advocacy. Baker's matter involved a self-represented claimant using AI assistance before Australia's workplace tribunal.
Access-to-justice limits remain
The Conversation's analysis places the ruling in the context of access to justice, particularly for individuals who cannot afford conventional representation. It also cautions that success with legal AI is likely to depend on user capability and careful oversight, rather than treating an AI agent as an autonomous substitute for legal expertise.
For legal-tech teams, the reported outcome is a concrete example of AI-supported self-representation in an employment classification dispute. It does not establish that AI-generated legal work is reliable across matters or users, but it adds to evidence that AI tools can help users identify legal options and navigate complex processes when they can validate the work.
Key Points
- 1The Fair Work Commission's ruling provides a documented example of AI-assisted self-representation succeeding in an Australian employment-status dispute.
- 2Baker reportedly used AI for issue spotting and case support, while the tribunal's legally binding decision remained a human adjudication.
- 3Comparable legal AI deployments can reduce research and drafting barriers, but reliable outcomes still depend on verification, procedure, and user capability.
Scoring Rationale
The ruling is a notable real-world example of AI assistance in a legally consequential self-represented employment dispute. It is relevant to legal AI builders and governance teams, although it is a single tribunal matter rather than a broad product release, precedent-setting AI regulation, or technical breakthrough.
Sources
Primary source and supporting public references used for this report.
Practice interview problems based on real data
1,625 SQL & Python problems across 15 industry datasets — the exact type of data you work with.
Try 250 free problems
