Garfield.Law Ltd, the first entirely artificial-intelligence-based firm authorised to provide regulated legal services in England and Wales, has recorded its first successful outcome in court. The technology assisted a claimant in recovering roughly £7,000 in unpaid fees at Wandsworth County Court in June 2026, little more than a year after the Solicitors Regulation Authority granted it approval. The result moves a closely watched experiment in legal automation from the regulatory pages onto the county court docket, and offers an early indication of how supervised AI systems may operate within the litigation process.
From regulatory approval to the courtroom
The SRA authorised Garfield.Law in May 2025, marking the first time a purely AI-based provider had been permitted to deliver regulated legal work under the regulator's oversight. The firm, founded by Philip Young and Daniel Long, concentrates on small-scale debt recovery and routine commercial disputes. Its June 2026 case saw the platform used by a claimant to pursue outstanding fees owed by a hospitality business, carrying a matter through to a favourable conclusion.
What the technology handled
According to accounts of the case, Garfield's system prepared pre-action correspondence, issued the claim, and produced trial documents, witness materials and court bundles. Human advocacy remained in the hands of counsel, with a barrister appearing on the matter. The division of labour illustrates the model on which the firm's authorisation rests: automation handles document-heavy procedural steps, while qualified professionals retain responsibility for representation and judgement.
Supervision and accountability
The regulatory framework does not remove human responsibility. Under the terms of its authorisation, solicitors remain accountable for the legal services delivered through the system, preserving the professional obligations that ordinarily attach to regulated work. That distinction matters at a time of heightened scrutiny over AI-generated litigation errors, including instances elsewhere of fabricated citations reaching the courts. Garfield's supervised structure is intended to keep procedural efficiency within established lines of accountability rather than replacing the oversight the profession is built upon.
An access-to-justice test case
The claim's modest value points to the space the firm is aiming at: disputes where conventional legal costs can approach or exceed the sum in issue, leaving smaller claimants without a practical route to recovery. Automating routine steps could lower that barrier, though the approach raises questions about how the market prices such work and whether clients will continue to pay traditional fees for tasks a supervised system can perform. The case sits against a wider backdrop of substantial investment in legal technology by established firms, and a profession still calibrating where automation belongs.
Garfield.Law's first courtroom success is a limited but meaningful data point. A single low-value recovery does not settle the broader debate over AI in legal services, yet it demonstrates that a regulated, supervised model can carry a claim from correspondence to judgment. As regulators, firms and clients weigh the balance between efficiency and accountability, the outcome will likely inform how far automation is trusted with contested matters, and how the boundaries of professional responsibility are drawn as the technology matures.
