AI-native UK law firm Keith raises £2 million in seed funding to launch a fully regulated conveyancing practice in Q3 2026, targeting a 70% reduction in transaction times
Keith, a UK-based AI-native law firm founded by Andy Shovel and Pete Sharman — the entrepreneurs behind plant-based food brand THIS — has raised £2 million in seed funding led by Backed VC, with participation from Breega and several angel investors. The capital will be used to build and launch a fully regulated AI law firm targeting residential property conveyancing — the legal process of transferring ownership of a property — as its first practice area. Keith plans to launch in Q3 2026 and will be regulated by the Council for Licensed Conveyancers (CLC), the specialist regulator for licensed conveyancers in England and Wales (as distinct from the Solicitors Regulation Authority, which regulates solicitors). This regulatory route allows Keith to offer conveyancing services without the full obligations of SRA authorisation, enabling a more streamlined operational model. The firm's core proposition is a 24/7 AI client agent and a target of reducing average conveyancing transaction times by 70%. The founders' logic is explicit: conveyancing is a large, fragmented market that has changed little in decades, making it structurally vulnerable to technology-led disruption — the same thesis they applied to plant-based food. Konnect's entry into the market represents a significant test of whether an AI-first legal service model can satisfy CLC regulatory requirements around client care, professional indemnity, and the duties owed to clients in what is typically their largest financial transaction. It also raises questions about where liability sits when an AI agent handles a conveyancing matter that goes wrong — a question that current CLC and SRA regulatory frameworks are not fully equipped to answer.
Why this matters
Keith's launch crystallises a regulatory question that UK legal regulators — the SRA, CLC, and ultimately the Legal Services Board (LSB) — will need to address: what obligations apply when a regulated legal service is delivered primarily or entirely by an AI agent, rather than a human lawyer? The firm's CLC authorisation route is deliberately chosen to reduce regulatory friction, but it does not resolve the underlying questions about professional duty, liability for AI errors, and client protection in high-stakes transactions. For law firms, the commercial threat is real but currently limited to commoditised, high-volume conveyancing — the same market that saw significant disruption from online providers. The broader significance is as a proof-of-concept: if Keith demonstrates that AI-native legal services can meet regulatory standards, the model will expand to other high-volume areas including employment, wills, and debt recovery.
On the Ground
A trainee working on the legal and regulatory side of an AI law firm launch would assist with reviewing the technology licence agreements governing the underlying AI model, marking up data processing agreements to ensure compliance with UK GDPR requirements for client data handled by the AI system, and preparing a regulatory impact assessment memo assessing CLC authorisation requirements and any conditions likely to be imposed.
Interview prep
Soundbite
Keith's CLC route lets it launch faster, but AI-delivered conveyancing exposes a gap in UK legal regulation that the LSB will eventually have to fill.
Question you might get
“If Keith's AI agent makes a mistake in a conveyancing transaction — for example, missing a restrictive covenant — who bears liability under the current regulatory framework, and is that framework adequate?”
Full answer
Keith has raised £2 million to build and launch an AI-native conveyancing law firm regulated by the CLC, targeting a 70% reduction in transaction times through a 24/7 AI client agent. The commercial logic is sound — residential conveyancing is slow, expensive, and ripe for technological disruption. The legal regulatory question is harder: neither the CLC nor the SRA has a clear framework for attributing liability when an AI system, rather than a human professional, makes the error that causes a client to lose money on a property transaction. This matters for the wider legal market because if Keith succeeds, the AI-native model will expand — and the regulatory gap will become impossible to ignore, likely triggering a formal LSB policy review that will affect every firm deploying AI in client-facing roles.
My notes
saved