AI helped a scientist sue his employer with no background in labour law

In a labour dispute with Macquarie University the computer specialist Greg Baker represented himself — and won the case at the Fair Work Commission. Before the process began his knowledge of labour law was, by his own account, practically nil.

The main conclusion here is not that AI has suddenly become a lawyer. Its more down-to-earth and perhaps more important role is to help a person decide to defend their own interests at all, when hiring a specialist is difficult or expensive.

That changes the balance of power not magically but organisationally. A person gets the chance to make sense of unfamiliar terms faster, to prepare questions and not to get lost in the procedure. For an employee in dispute with a large employer, that alone can be a substantial advantage.

But a loud headline easily over-promises. It is not clear from the material which model or service Baker used, which documents he prepared with AI's help, or what precisely brought the win. The result cannot automatically be treated as proof of neural networks' legal reliability: an error in a deadline, a provision or a reading of the facts can damage a case rather than help it.

The practical boundary looks like this: AI can be a navigator through an unfamiliar system, but responsibility for decisions stays with the person. Would you trust a neural network to prepare your case in a dispute with an employer if you could not afford a lawyer? Source: Hacker News - Newest: ""AI" "LLM""