Adoption of artificial intelligence for personal finance is surging. More than half (55%) of Americans said they used AI to help manage finances, according to a February 2026 survey conducted for TD Bank. That was up from only 10% a year before.

I know our profession is tempted to read a finding like that as a threat. I suggest curiosity instead. People are exploring, pressure testing, asking questions at midnight that they might never voice in a conference room. Advisors should expect that, and pay attention to it. What's driving those questions, especially for preretirees, is usually just a hunger for information. It's an underlying anxiety.

The anxiety was already there. In our research at Edelman Financial Engines on financial confidence, we found a paradox. As people approach retirement, their moderate level of self-reported confidence holds steady-while aspects of their financial anxiety climb sharply. What struck me is that more than a third of adults ages 55-64 cite "second-guessing myself" as a major driver of financial anxiety, up significantly from younger age groups.

I have spent my career studying the cross lines of human behavior and financial decision-making. In that time, I've seen a consistent undercurrent: anticipated regret-the fear of looking back years from now and realizing you chose poorly. At age 35, many financial mistakes are recoverable, with decades to course-correct. By the late 50s, the decisions your clients face-when to retire, when to claim, how much to draw, how much risk to carry-feel permanent, interconnected, and hard to reverse. None has a clean "right" answer.

Your clients aren't anxious because they lack information. They're afraid of regretting decisions they haven't even made yet. And now that population has a machine that seemingly answers any financial question, instantly, at 2 a.m.

Where it falls flat. AI is useful in so many ways. It can lay out claiming strategies, model withdrawal scenarios, and explain Medicare trade-offs in seconds. But for an anxious preretiree, two AI habits create serious issues.

First, volume. Ask an AI platform about Social Security, and you'll receive 10 claiming strategies, seven withdrawal approaches, and three alternative retirement dates. Decades of research on choice overload tell us what happens next: as options multiply, people come less likely to act and less satisfied with the choices they make.

Second, agreement. When an overwhelmed person leans toward a direction, the platform tends to concur. A Stanford study published in Science this year found that leading AI models endorsed a user's position significantly more than humans did-and that people trusted the agreeable models more. The answers arrive confident either way: Nearly four in 10 Americans say they'd feel comfortable acting on generative AI's financial advice without verifying it anywhere else, according to a 2025 CFP Board consumer survey.

Put those together and you have a tool that amplifies exactly what drives retirement anxiety: It multiplies a person's options, then validates whichever direction they already lean-all delivered with total confidence. Not because the technology is malicious, but because it isn't built to push back. For a 35-year-old, that is a recoverable detour. For someone facing a permanent decision about when to claim Social Security, it's a risk that compounds.

The opportunity: lean in with urgency. Today's heaviest AI users skew younger and high-income-preretirees are still the lightest adopters-with adoption stepping down with age from Gen Z to millennials to Gen X to boomers, according to TD Bank. This isn't a reprieve, but a window to be prepared as advisors.

The immediate task at hand for all advisors: get curious. Lean in. Ask your clients what they have been asking their favorite AI model. Don't police it-use it. Their prompts are a map of their worries and what matters to them, and every question they took to AI in the middle of the night is an invitation to a conversation they didn't know how to start with you. A client who shows up holding three conflicting AI strategies is an invitation for you to show your true value, as a human.

The challenge isn't to compete with AI on answers. It's to provide the thing anxious clients can't get from it: human judgment based on context that machines don't have. Which choices matter most to this client? Which trade-offs can they live with? When do they need an ear for an emotional conversation? AI doesn't know what's uniquely important to your client beyond whatever it is fed: media, views, clicks, pictures, reviews, websites, locations, purchases, facts, figures. Does our digital footprint determine who we are? I don't think so.

If retirement anxiety were an information problem, AI would have solved it by now. It's a regret problem-and regret isn't answered with information. It's answered by someone who has been there, who has helped other people navigate the complexity of life's financial decisions. AI hasn't replaced that. Ultimately, AI will remind millions of anxious people, one overly confident answer at a time, exactly what it can and can't give them. So be there when clients (and prospective clients) come looking for you.

Michael Liersch is Chief Planning Officer at Edelman Financial Engines, where he leads the firm's strategy for financial planning, advice, and client solutions. A recognized expert in behavioral finance and wealth planning, he has held senior leadership roles at several major financial institutions and is known for applying cognitive science to financial decision-making. Michael holds a PhD in cognitive psychology from the University of California, San Diego, and a bachelor's degree in economics from Harvard University.

Editor's note: Guest commentaries like this one are written by authors outside the Barron's Advisor newsroom. They reflect the perspective and opinions of the authors. Submit feedback and commentary pitches to advisor.editors@barrons.com.