Australia's Fair Work Commission ruled on August 12 that Macquarie University computing academic Gregory Baker should be treated as an ongoing part-time employee. Baker represented himself after using AI tools to identify a possible casual-to-permanent employment pathway and support his case, according to The Conversation, but the legally binding decision was made by the human tribunal.
Australia's Fair Work Commission ruled on August 12, 2026 that Gregory David Baker, a computing academic at Macquarie University, should be treated as an ongoing part-time employee. The published decision is Application by Gregory David Baker to deal with a dispute about changing from casual employment, [2026] FWC 3054.
Baker represented himself. According to The Conversation, he used AI tools to identify the employee-choice pathway that allows eligible casual workers to seek conversion to full-time or part-time status and to help organize the dispute. He had taken the matter to the Commission in December 2025 after the parties could not resolve it directly.
What the ruling establishes
The Commission's decision concerns Baker's employment status under Australia's casual-employment framework. It is not a ruling that an AI system can act as a lawyer, and the AI did not decide the case. Baker remained responsible for the application and submissions, while the Commission assessed the evidence and made the binding determination.
That distinction matters because accounts of "AI winning a case" can blur three separate roles: finding a potentially relevant legal pathway, helping a person research or prepare material, and formally representing or adjudicating a dispute. The documented event supports the first two uses. It does not transfer professional responsibility or judicial authority to the software.
Access to justice, with limits
The case is a concrete example of AI reducing the research and drafting burden for a self-represented person in a legally consequential matter. It also shows why legal AI needs strong verification. A plausible answer can still cite the wrong jurisdiction, miss a deadline, misunderstand evidence, or invent authority. Those failures can be costly even when a tool is useful for issue spotting.
For builders and institutions, the practical design goal is assisted navigation rather than automated certainty: link claims to retrievable primary law, preserve the user's source trail, flag jurisdiction and date, and make clear when professional advice may be necessary. One successful tribunal matter does not establish a general success rate for AI-assisted litigants, but it does provide a traceable public example of the technology helping a person reach and argue a valid legal pathway.
Key Points #
- 1The Fair Work Commission's August 12 decision in [2026] FWC 3054 treated Gregory Baker as an ongoing part-time employee.
- 2Baker used AI for issue spotting and case preparation, while he remained self-represented and the human Commission made the binding decision.
- 3The case shows access-to-justice potential but does not establish that AI is a lawyer or that AI-assisted litigants generally succeed.
Scoring Rationale #
The decision is a documented real-world use of AI assistance in a consequential employment dispute and has direct relevance to legal-product design and access to justice. Its significance is bounded because it is one tribunal matter, not a broad precedent on AI representation.
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.