A solicitor in a regional practice is acting for a tenant in a commercial lease dispute. The lease contains a clause that looks unreasonable on its face: an indemnity shifting all liability to the tenant regardless of fault, coupled with a termination right letting the landlord walk away on seven days' notice. The question is whether this amounts to an unfair term under the Australian Consumer Law.
The firm's practice management system holds everything else: the matter file, the trust accounting, the correspondence, the time records. Whether it is Leap, Actionstep, Smokeball, or one of the older incumbents, that decision was made years ago and the firm has no reason to revisit it. The PMS runs the operational backbone of the practice. But when the solicitor needs to research whether this lease clause is an unfair term, the PMS offers nothing. It was never built to.
The obvious next step is a general-purpose chatbot. What comes back is fluent, confident, and wrong. The citations do not exist. The principles are drawn from American or English authority that has no application in an Australian court. The surface plausibility collapses the moment you try to verify the answer against a real report. For a sole practitioner in Albury or a three-partner firm in Mackay, this is not a minor inconvenience. You cannot advise a client on the strength of a citation you cannot verify, and you cannot file submissions built on authorities that dissolve under scrutiny. The time you spend checking whether the AI's answer is real is time you might as well have spent running the research the old way, through the databases you already trust.
Practice Note SC Gen 23 in New South Wales, which took effect on 3 February 2025, makes the stakes explicit. Legal practitioners using generative AI must verify every legal citation, case reference, and legislative provision, and that verification cannot itself be delegated to the AI. The obligation sits with the practitioner. Generic AI tools were built to produce text, and the professional obligation to verify every citation sits outside their design.
The solicitor opens Habeas in a browser tab and types the question in plain language, close to what they would ask a colleague. Habeas' Search Engine scans over 300,000 Australian cases and pieces of legislation in seconds. The results are grounded in a closed dataset of legitimate Australian legal sources, so they are verifiable and traceable, never hallucinated. Each result carries the case name, the citation, and a pinpoint to the paragraph where the relevant principle sits. The solicitor clicks through to the judgment and reads the passage in context, not a paraphrase of it.
Authorities are visible and checkable. The solicitor can see which cases the platform has relied on, verify them against the source, and form a preliminary view built on real law. Where the matter involves a discovered set or a brief bundle, Document Stores lets the firm upload those materials and query them directly, grounding answers in the firm's own documents alongside the public authorities.
Regional practitioners work without the support infrastructure of a large firm. They have no law library down the corridor and no junior solicitor to delegate to. The closest law society library might be a two-hour drive. A specialist database subscription might be beyond the budget. Running a semantic search across Australian primary law from a desk in Wagga Wagga or Burnie, and getting back something you can rely on, changes the kinds of matters a small practice can take on with confidence.
The judgment remains the solicitor's. We built Habeas to return grounded, cited results from Australian primary law, but the analysis of those results is the practitioner's work. Under SC Gen 23 and equivalent professional obligations across jurisdictions, the duty to verify sits with the solicitor. Habeas supports that duty by making verification straightforward, because every citation links to its source. The platform gives you the authorities. You draw the conclusions and stand behind the advice you give.
Foundational research processes that used to take a full morning can now be completed in minutes. A sole practitioner who can resolve a research question at that speed can take on matters that would previously have been uneconomic. A three-partner firm that can offer clients the depth of research associated with a larger outfit, without the overhead, competes for work that would once have gone to a city firm. The PMS records the time. Habeas handled the research. The solicitor gets back to the work that requires a solicitor.
If your firm is working through how to add research depth without disrupting the operational backbone you have built, see what Habeas does at habeas.ai. Book a demo, run a query on a matter you are working on, and check the citations yourself. The proof is in the sources.
If you want to try for yourself or get in contact, book a demo with us here. We also offer the capacity for self-serve individuals to sign up, and subscribe or register a free trial at app.habeas.ai.
The legal research in this article was conducted and every citation verified using Habeas, the Australian legal AI research platform.
Hero image: Virginia Chien on Pexels
