Small business owners are, by nature, a self-sufficient breed. They fix problems, find solutions, and get things done. So, when AI tools arrived promising instant answers, ready-made documents and round-the-clock availability, the uptake was inevitable.
Yulia Barnes, Managing Partner at Barnes Law, explains why the rise of AI is making getting proper professional legal advice more important than ever, and why AI is only as good as the data it’s fed.
The draw is not simply cost. For most business owners, the real appeal is speed and control. AI feels like the answer to both problems. It is immediate, consistent, and it never pushes back. The difficulty, of course, is what it actually produces.
Here is an analogy that I think captures it well. Imagine a person who has read everything, every contract ever 
drafted, every piece of legislation, every legal textbook in print. They can discuss any of it with apparent fluency. They sound, frankly, impressive. Ask them to apply any of that to your specific situation, however, and the wheels come off. The output is sophisticated in appearance and useless in practice. It does not know your business, your counterparty, your regulatory position, or what you are actually trying to achieve. That is AI. Well-read to a fault and unable to do the one thing that matters, exercise judgement.
The text AI produces tends to compound the problem. It is dense, technical, and carries the tone of authority. Most people reading an AI-generated content would have no reason to question it, but scrutinise it closely, or try to rely on it when something goes wrong, and the gaps become obvious. Clauses that appeared watertight turn out to be circular or contradictory. The language sounds like law but does not behave like it. It is constructed to read as correct. That is not the same thing as being correct, and in legal practice, the difference between the two can be extremely costly.
What AI cannot replicate is context. Every legal matter sits within a broader picture: the nature of the relationship, the commercial objectives, the regulatory landscape, the risk profile of the business. When a client asks a legal question, part of the answer involves recognising what has not been asked but needs to be. That judgment comes from experience, not data. AI answers the question in front of it. It has no visibility of the question behind it, and that is precisely where most problems originate.
There is also a point here that the legal profession does not always make clearly enough. AI can be a useful tool in the right hands. A senior lawyer with fifteen years of experience can interrogate AI output, identify where it falls short, and apply the necessary judgment to make it useful. A business owner with no legal background cannot do the same. They do not have the frame of reference to know what is missing. This is also why, in my own practice, I do not permit junior lawyers to rely on AI. Legal judgment is built through experience, through doing the work, sitting with difficult problems, and developing the instinct to know when something is not right. There is no shortcut to that. A client using AI in place of a lawyer is not just cutting costs. They are operating without the judgment that makes legal advice worth having.
The consequences are rarely immediate, which is part of what makes this so dangerous. A poorly drafted clause does not announce itself. It sits in an agreement quietly, sometimes for years, until the circumstances arise that expose it. By that point the dispute is live, the costs are significant, and the options are limited. The original saving tends to look rather different in that light.
One further point that tends to go unmentioned. AI has no professional indemnity insurance. Every qualified solicitor does. If a lawyer gets it wrong, there is regulatory accountability, a complaints process, and insurance to meet the consequences. If AI gets it wrong, the liability sits with whoever relied on it. There is no claim to bring, no regulator to call, and no redress. Just a document that did not do what it needed to do, and a business owner left to manage the fallout alone.
Good legal advice does more than prevent mistakes. It ensures documentation is accurate and enforceable. It keeps a business compliant across employment law, data protection, commercial contracts, intellectual property and sector-specific regulation. It provides the kind of considered, commercially-aware guidance that supports sound decisions at every stage of growth. The businesses that understand this treat legal counsel as a core part of how they operate. The ones that do not tend to find out why it matters at the point when it is most difficult to fix.
AI will continue to develop, and its role in legal services will evolve with it. That is not the argument here. The argument is simpler. For any business owner navigating real legal complexity, contracts, compliance, commercial risk, there is no substitute for qualified professional advice. The more convincingly AI presents its output, the more important it becomes to know when not to rely on it and that, in the end, is a question only a lawyer can answer.
Yulia Barnes, Managing Partner at Barnes Law,