Listen to this articleArtificial intelligence-powered apps have the potential to provide low-cost financial advice to the masses. Yet, a new study finds AI only delivers correct answers about half the time.
According to a recent College Investor report, Google AI overviews appear misleading or inaccurate in 43% of finance-related searches.
The report’s authors warn this could confuse consumers, claiming AI-generated content is particularly weak at tax and financial-aid-related queries.
The tech’s poor accuracy hasn’t diminished its appeal.
According to a recent poll from Ipsos, almost 2 in 5 Americans (37%) use AI to manage their finances. The adoption rate is even higher among younger consumers; 61% of Gen Zers link the technology to their money.
Many AI web apps are free or relatively inexpensive to use. Yet, when it comes to money, inaccuracy can do more than waste your time; it can harm your financial health, too. How do (human) financial advisors view AI usage in their industry?
Right, Wrong, and Between?
AI is notoriously disorienting. One minute, it seems to work magic; the next, it fails hard. Users have all sorts of odd experiences with the technology, and its accuracy can be hard to predict — sometimes spot on and, at other times, wildly off.
The researchers at College Investor were perplexed to find this dichotomy with Google AI Overviews. They found it served up outdated information on straightforward topics like student loan repayment plans and CD rates.
Nevertheless, they showed a firm grasp of trending topics, such as the ” Chase Glitch,” a recent vulnerability in Chase Bank’s system that enabled overdraws.
The report noted that AI answers were only partially correct, missing vital information that could cause financial harm.
The Internet has been an open source of financial knowledge since its inception, and when used correctly, it enables the democratization of financial education.
Historically, Google Search helped people navigate the personal finance blogosphere and sort good social media content from the bad. However, with the fast adoption of AI platforms like ChatGPT and Microsoft Copilot, users go straight to these generative tools for answers.
Financial advisors see the effects.
“Clients have come to me multiple times with incorrect information assembled by AI,” says Benjamin Simerly, Founder of Lakehouse Family Wealth. “When fact-checking info with clients, I emphasize that AI tends to work as an aggregator, not an original source. More often than not, the “popular” websites do not necessarily put the most effort into accuracy. So, the best answer is almost always the original source. For instance, for tax questions, that’s the IRS website.”
Although he tries to steer people in the right direction, Simerly understands the appeal of these content tools.
“The most accurate sites, generally aren’t the most sexy,” he laments. “After all, who wants to read serious financial and legal documents over an influencer’s fun video?”
Yet there is a severe trade-off, especially when questions concern serious financial decisions.
“I’ve seen AI routinely get the numbers wrong for various retirement account contribution limits,” says David Nash, Founder of Tend Wealth. “The IRS limits are so important, and with bad data, it’s easy to run into a painful tax issue.”
Nash contends AI struggles to keep up with the ever-changing numbers needed for proper financial planning, which means advisors must still rely on their training to serve clients effectively.
It shouldn’t surprise anyone that AI is generic, not nuanced. The last three letters of ChatGPT stand for General-Purpose Technology.
“AI tends to be generalized, and personal finance is so specific to each individual’s circumstances that there’s really no substitute for personalized advice from someone who knows your situation,” says Jason Gilbert, Founder and Managing Partner of RGA Investment Advisors. “I usually recommend clients check multiple reliable sources or, better yet, talk to their advisor directly when they come across something concerning.”
Looking to the Future
The consensus is clear for now: the cost of hiring a financial advisor with real-life experience offers more value than AI tools.
Artificial intelligence often struggles with nuanced understanding and emotional intelligence, essential for effective communication. Additionally, as net worth by age grows for most individuals and couples, so does the complexity of financial planning.
AI can often misinterpret and fail to personalize content for individuals’ needs, but that may not always be the case.
According to one predictive forecast by Gartner, by 2027, more than 50% of the GenAI models that enterprises use will be specific to their industry or business function — a radical surge forward from approximately just 1% last year.
“We don’t yet have AI tools we can trust on financial matters,” says Simmerly. “That will come in time, though, we are sure.”
Mitlin Financial Founder and Wealth Advisor Lawrence Sprung succinctly sums it up for consumers: “Checking with a professional before taking action in your financial life based on AI is highly recommended.”
Consumers and professionals must adapt as AI becomes more integrated into personal finance ecosystems. Yet the data suggests it is not time to jump on the AI bandwagon. For now, reliable primary sources and human judgment are essential. In the future, as technology evolves, combining AI tools with human expertise could elevate financial know-how further.
This article was produced by Media Decision and syndicated by Wealth of Geeks.
(Except for the headline, this story has not been edited by PostX News and is published from a syndicated feed.)
