AI wills looked fine. SE Solicitors found them missing inheritance tax and other key questions

A UK probate firm put an AI chatbot on the stand to draft a will for a fictional client named “Daniel,” and the result looked good until it was interrogated. SE Solicitors says the document read well enough, but it missed many of the questions and safeguards a probate solicitor would normally cover before anything becomes legally binding.

SE Solicitors senior associate Tom McInerney reviewed the chatbot’s draft and warned that it “missed many of the questions” that matter. According to the firm, the chatbot did not attempt to explore inheritance tax planning, lifetime gifts, pensions, life insurance, digital assets, trusts, or whether either child might be financially vulnerable. It also failed to ask whether Daniel had a partner, whether he wanted to leave gifts to friends or charities, or whether he wished to exclude anyone from inheriting.

That contrast is the whole story, and it should worry anyone who thinks AI is ready to “replace” the legal middle. A will is not a template you fill out and forget. It is a legally binding instrument designed to survive real humans, real assets, and real disputes. In McInerney’s words, “a will is not a letter or a cover note, it is a legally binding instrument that must hold up in court.” The firm’s point is blunt: AI can produce text that fits the prompt, but it cannot reliably invent or uncover the missing context that would normally be gathered through a solicitor’s questions.

SE Solicitors argues this is structural. The chatbot could only work with what it was given. A solicitor, by contrast, typically spends time asking follow-up questions before drafting a will. “AI can't see the gaps because it has no idea about what you didn't tell it,” McInerney said. “All it can ever produce is a will that fits the prompt, not the person.” That distinction matters because probate work is not just drafting. It is a guided fact-finding process that surfaces the tax, family, and asset details people often do not know are relevant until someone asks.

The firm also criticized the drafting mechanics themselves. In one clause about the family dog, the chatbot left a placeholder instead of specifying how much money should be set aside for its care. It also failed to make that gift conditional on the named beneficiary actually taking the animal. Those may sound like “small” errors, but in estate planning small errors can translate into confusion or disputes later, especially when the person who made the will is no longer around to clarify intent.

SE Solicitors also raised second-order risks tied to changing circumstances and regulatory complexity. It warned that relying on AI could cause consumers to overlook changes in inheritance tax rules or the need to review a will as circumstances change due to events like marriage, divorce, having children, property purchases, or other major life events. That warning connects to a broader pattern: estate planning is not one-and-done paperwork. It is a living decision that often gets revisited as laws, assets, and family structure evolve.

On data privacy, the firm added a second layer of concern: anyone who asks an AI platform to draft a will is likely handing over sensitive personal and financial information. SE Solicitors argued consumers may not fully understand how that data is stored or processed, particularly compared with professional confidentiality obligations placed on solicitors. For decision-makers watching AI creep into regulated services, this is the recurring theme: even if AI can generate an artifact, the process around it, including confidentiality and governance, can lag far behind what clients assume they are buying.

And yet, the demand signals are real. SE Solicitors cited Google Trends data showing searches for “legal AI” have risen 312 percent over the past year. Separately, research commissioned by the Association of Lifetime Lawyers found 72 percent of UK adults aged 30 to 34 would consider using AI to write or update their will. The experiment suggests AI is currently good at converting provided information into will-like text, not at doing the missing diligence of asking the questions that determine whether the document actually fits the person and stands up in court.

For boards, investors, and operators building or buying into legal AI tools, the takeaway is not “stop using AI.” It is that the risk profile may be different than people expect. If AI becomes part of consumer legal workflows, the failure mode shifts from “blank page” to “plausible output with invisible omissions.” That is a product liability and trust problem, not just a technology problem. And for firms in probate, the opening is also clear: the defensible work is the human-led questioning, the governance, and the safeguards that turn a prompt into something that survives reality.