The Financial Times highlights a booming, chaotic legal AI market. Before you sign another SaaS contract, look at the foundations underneath your firm.
If you read the Financial Times’ latest deep dive into legal technology, you could be forgiven for feeling a spike in blood pressure.
According to the report, there are now more than 5,000 legal tech companies vying for your firm’s attention. From hyper-specialized deal-sheet extractors to AI “digital twins” of retiring partners, the market has moved rapidly from general chatbots to aggressive, niche automation.
For Magic Circle firms with dedicated innovation teams and bottomless R&D budgets, testing eight different early-stage start-ups is part of the game.
For the UK mid-market law firm, however, this explosion creates a very real operational hazard: tool fatigue, fragmented data, and unquantified risk.
When we speak with Managing Partners and COOs across the UK, the anxiety isn’t about missing out on the latest Silicon Valley start-up. It’s about buying expensive software that lawyers don’t use, or worse, accidentally compromising client confidentiality in the rush to modernize.
Buried beneath the start-up profiles in the FT piece are three truths that every mid-market firm should use as a filter before spending another penny on AI:
1. Your Legacy Infrastructure Will Dictate Your AI ROI
The FT notes a critical caveat: firms must ensure legacy IT infrastructure does not conflict with or undermine new legal software.
Many of the emerging tools highlighted—like DeepJudge or Twin1 AI—rely on connecting directly to your Document Management System (DMS), emails, and Teams channels to surface institutional knowledge. If your underlying data is unstructured, trapped in legacy on-premise servers, or lacks clear permissioning, plugging in an AI search tool won’t give you clarity. It will simply surface the wrong documents faster—and potentially expose sensitive matter files across ethical walls. Data governance isn’t a nice-to-have precursor to AI; it is the whole ballgame.
2. Don’t Buy Software to Fix a Pricing Problem
One of the sharpest observations in the report comes from Centari’s founder: “The law firm business model is still built for the pre-AI era, and they’re plugging AI into the system.”
If a tool cuts due diligence or side-letter drafting time by two-thirds, and you are still billing strictly by the hour, you haven’t increased profitability—you’ve just shrunk your revenue. Mid-market firms don’t need to rush out and buy early-stage US start-ups to solve this. Start by asking a simpler commercial question: Where are we losing unbilled time to administrative friction, and how do we move those specific matters toward fixed-fee or value-based pricing?
3. Adoption Fails in the Classroom, Not the Server Room
Northwestern’s Daniel Linna puts it bluntly: “The biggest risk now is not buying the wrong legal tech; it’s failing to educate your workforce.”
Before adding a 15th login to your fee-earners’ desktops, look at the tools you already pay for. Most mid-market firms are utilising a fraction of the security, automation, and AI capabilities already sitting inside their Microsoft 365 and core practice management environments. Empowering fee-earners with structured, SRA-compliant training on existing platforms delivers immediate, risk-free productivity gains without adding another vendor to your attack surface.
The Mid-Market Advantage
You don’t need to place speculative bets on 5,000 start-ups. Let the global giants fund the R&D and iron out the compliance wrinkles over the next 24 months.
Your competitive advantage right now lies in ruthless operational simplicity: lock down your data governance, modernize your core cloud architecture, train your people on the tools they already have, and let the market consolidate around you.