AI chatbots deliver confident financial advice that's often wrong, leaving investors burned

AI Chatbots Gain Traction in Personal Finance Despite Growing Concerns

AI chatbots have rapidly infiltrated personal finance decision-making, with nearly a fifth of UK consumers now using AI for financial guidance, according to a Financial Conduct Authority report 2

2

. In the U.S., adoption rates tell a similar story: a 2025 Pew Research Center survey found 34% of U.S. adults and 58% of those under 30 have used ChatGPT, roughly double the share from two years earlier 1

1

. This shift toward AI-powered financial advisers marks a significant change in how people approach money management, but the consequences are starting to emerge.

The dangers of relying on AI chatbots for financial decisions are no longer theoretical. According to a 2025 survey of 2,000 U.S. adults by Pearl.com, 19% said they lost more than $100 by following AI financial advice from a chatbot 3

3

. Among Gen Z investors, that figure climbed to 27%, highlighting how younger users—despite their digital fluency—are particularly vulnerable to AI's confident but flawed recommendations 5

5

. These losses represent only what people have noticed; the actual toll may be far higher.

Why AI Is Giving People Bad Money Advice

The core problem with relying on AI for personal finance advice lies in what finance experts call the "fluency trap." AI chatbots deliver answers that sound authoritative and well-organized, creating an illusion of competence that masks fundamental flaws. A finance professor watching this trend warns that fluency is not accuracy—a chatbot can be word-perfect and still provide wrong guidance about taxes because it never asked about crucial details 1

1

.

Consider retirement savings decisions: an AI might recommend when to claim Social Security and how to convert retirement accounts without considering that a spouse's age and health status could completely flip the math, or that the suggested conversion would trigger higher Medicare premiums years later 3

3

. These quiet failures are far more dangerous than obvious mistakes because they prevent people from seeking professional help when they need it most.

Research published in the Journal of Financial Planning examined seven widely available generative AI platforms—including ChatGPT, Claude, Copilot, DeepSeek, Gemini, Meta AI, and Perplexity—and found "significant variation" in how they answered prompts about emergency savings, asset allocation, and retirement portfolio withdrawals 4

4

. The study revealed that GenAI-driven responses "may sound confident but can still be incomplete, misleading, or incorrect," raising serious questions about consistency and fairness.

The Risks of AI-Generated Financial Advice Intensify Where Stakes Are Highest

AI's role in financial decision-making becomes most problematic precisely where people need help most. While AI chatbots handle routine topics well—explaining what a Roth IRA is or how compound interest works—they struggle with rare, complicated, one-time decisions like exercising stock options, understanding the alternative minimum tax, or developing Social Security strategies for couples 1

1

.

Researchers describe this uneven competence as a "jagged frontier"—reliable with common cases but unreliable for unusual ones. In finance, the unusual cases tend to be the expensive ones 5

5

. Financial guidance operates as what economists call a credence good, like a mechanic's diagnosis or doctor's recommendation. You often cannot tell whether the advice was sound, sometimes for years. A mistaken tax move may not surface until an audit; a flawed 401(k) drawdown plan may not cause problems until the stock market slumps 3

3

.

Andrew Lo, director of MIT's Laboratory for Financial Engineering, emphasizes the concern: "One of the things about LLMs that I find particularly concerning is that no matter what you ask it, it'll always come back with an answer that sounds authoritative, even if it's not" 4

4

. These hallucinations—wrong answers delivered with complete confidence—create a dangerous situation where users have no way to detect errors until significant damage occurs.

Regulatory Gaps Create Uneven Playing Field for Wealth Managers

The rapid adoption of AI-powered financial advisers has exposed a regulatory gap that puts traditional wealth managers at a disadvantage. Established financial firms face strict rules about providing personalized advice and face punishments for errors, while tech platforms offering AI chatbots operate without such constraints 2

2

. Banks like Lloyds and Barclays are developing tools to provide "targeted support"—a midpoint between specific advice and generic guidance—but must carefully calibrate their offerings to comply with regulations.

Meanwhile, Google's Gemini will confidently tell a UK saver with minimal personal information that "your absolute priority should be a Lifetime ISA," without the regulatory oversight traditional advisers face 2

2

. Only 40% of respondents in the FCA survey realized there was no way to complain if something goes wrong after consulting AI about their finances. The Financial Conduct Authority has noted the risk of an "uneven playing field" between regulated firms and tech platforms, though swift action remains unlikely.

Crucially, AI lacks fiduciary duty to users, meaning it doesn't legally need to provide financial advice in users' best interests 4

4

. This fundamental difference between AI chatbots and human advisers creates a protection gap that could leave consumers vulnerable to suboptimal or biased outputs for years before problems become apparent. As lost money following AI financial advice continues to mount, particularly among younger investors, the pressure for regulatory intervention will likely intensify—though whether that happens before more people suffer losses remains uncertain.