A growing number of Americans are turning to AI chatbots for financial guidance, and researchers warn that the confident tone of these tools may be leading users to make costly mistakes. According to a 2025 survey by Pearl.com, a professional services platform, 19% of U.S. adults who used an AI chatbot for financial advice reported losing more than $100 as a result. Among Gen Z investors specifically, that figure rose to 27%. The findings come as usage rates have roughly doubled in two years, with 34% of all U.S. adults and 58% of those under 30 now having used ChatGPT, according to Pew Research Center data.
The research is drawing attention to what finance professor Argun Saatcioglu calls the 'dangerous failure' mode of AI financial advice: when a chatbot delivers a polished, confident answer that is simply wrong. Unlike obvious errors that trigger users to seek human help, these quiet failures often go undetected for years because people lack the expertise to evaluate the guidance they received.
What the Left Is Saying
Consumer advocates and progressive policy voices say the findings highlight the need for greater oversight of AI tools in financial services. They argue that without guardrails, companies have little incentive to clearly communicate the limitations of their products. Consumer Financial Protection Bureau guidelines recommend verifying any adviser's credentials and ensuring fee transparency, but critics note these standards were designed for human advisors and may not adequately address automated systems.
Senator Elizabeth Warren has previously called for expanded CFPB authority over fintech products, arguing that algorithmic advice tools should face the same disclosure requirements as traditional financial advisors. Consumer groups contend that vulnerable populations—including retirees managing fixed incomes—are disproportionately at risk when AI guidance goes awry without any warning signals to prompt course correction.
What the Right Is Saying
Free-market advocates and tech industry representatives argue that prescriptive regulation could stifle innovation in beneficial financial technology. They maintain that users bear responsibility for exercising judgment about which decisions warrant professional advice versus general information gathering.
The Chamber of Commerce has cautioned against regulatory approaches that would limit consumer access to free or low-cost AI tools, arguing that the technology serves as a democratizing force for financial literacy. Conservative commentators have emphasized individual choice, noting that AI chatbots can provide valuable baseline education on topics like compound interest or account types without the cost barriers of traditional advisors. They contend that the solution is improved user discernment rather than restricting access to information.
What the Numbers Show
The survey data reveals significant financial exposure from AI-assisted decisions: 19% of respondents who followed chatbot advice reported losses exceeding $100, rising to more than one-quarter among Gen Z investors. The Pew Research Center figures show ChatGPT adoption has grown to 34% of U.S. adults overall and 58% among those under 30—roughly double the rates from two years prior. Research on robo-advising platforms indicates their users tend to be younger, predominantly male, smaller retail investors who increase activity during periods of market volatility—when wrong moves carry the highest cost.
Finance researchers describe AI's uneven reliability as a 'jagged frontier,' where performance is strongest on routine questions like IRA mechanics but weakest on complex one-time decisions involving taxes, estate planning, or Social Security claiming strategies. Experts note that financial advice functions as what economists call a 'credence good'—meaning clients often cannot evaluate whether the guidance was appropriate, sometimes for years after receiving it.
The Bottom Line
The research suggests the primary danger may not be acting on bad AI advice but rather never seeking human guidance at all when circumstances warrant it. As AI chatbots become more polished in their responses, experts caution users to treat them as educational starting points rather than definitive answers—particularly for decisions involving large sums, tax consequences, irreversible actions, or situation-specific factors. Red flags indicating AI may be operating beyond its reliable scope include estate questions, retirement drawdown planning, Social Security claiming strategies, and major one-time transactions. Wealth management firms are already facing what industry observers describe as a 'chatbot reckoning,' with investors factoring in potential disruption from automated advice tools into company valuations.
The broader policy question centers on whether existing regulatory frameworks adequately protect consumers from confident-but-wrong AI guidance—or whether new disclosure requirements and accountability standards are needed for financial technology products.