AI threatening financial advisors: MIT professor points to a major obstacle

X Facebook WhatsApp

More Americans are turning to generative artificial intelligence for money matters, but the tools that offer investment and tax guidance still sit in a legal gray area — and that gap matters now because millions are already acting on AI-backed recommendations. As the technology improves, experts say consumers could gain convenience while losing the protections they’d have with a human advisor.

Finance scholars and legal researchers who study the intersection of AI and financial services acknowledge the technology’s growing capability to analyze markets, model scenarios and explain complex products. Yet they warn the key difference between an algorithm and a licensed professional is not sophistication — it is legal accountability.

AI can advise — but it doesn’t carry a legal duty

At institutions like MIT Sloan, academics note that large language models and related systems have developed substantial financial knowledge. Still, those systems do not operate under a fiduciary duty — the legal obligation that requires many financial professionals to put a client’s interests ahead of their own.

That absence of a binding duty matters because advisors who breach fiduciary standards can face regulatory action, civil suits or even criminal penalties. Without comparable legal exposure, AI providers lack the same direct incentive to avoid conflicts of interest or costly errors, experts say.

Use is widespread — and growing

A September Intuit Credit Karma survey found that roughly two-thirds of Americans who have tried generative AI have used it for financial advice, a proportion that rises to more than four in five among younger adults. The poll also reported that about 85% of those users followed at least some of the suggestions they received. That level of reliance raises immediate regulatory and consumer-protection questions.

“People are increasingly getting concrete recommendations from these services, and it’s unclear who is accountable when things go wrong,” said a senior research fellow at NYU Law’s Information Law Institute, pointing to an unresolved legal picture around corporate responsibility for AI-driven advice.

Where AI helps — and where it can mislead

There are practical applications where AI already adds value. Models can quickly summarize complex topics like Medicare enrollment rules or lay out the trade-offs between account types, giving consumers a clearer starting point for decisions.

But experts caution against treating algorithmic responses as definitive financial plans. For personalized, number-driven tasks — calculating tax liabilities, projecting retirement income, or deciding whether to roll over a 401(k) — AI can be unreliable. Large language models tend to produce answers that sound confident even when they are incorrect, making independent verification essential.

Useful: Broad explanations, scenario generation, initial education on financial topics.

Risky: Specific calculations, tax filings, individualized investment allocations without human oversight.

Action to take: Double- or triple-check AI outputs and consult licensed professionals for major or complex financial moves.

A representative from a private AI firm publicly reminded users that the product is “not tax advice,” underscoring the industry’s own hedging around liability and accuracy.

Not every human advisor offers the same legal protection

Complicating the picture further, not all human financial professionals are fiduciaries. The rules that govern brokers, registered investment advisors, insurance agents and other intermediaries differ, and those distinctions shape what legal obligations, if any, apply to their recommendations.

Regulatory efforts to expand fiduciary duties have faced setbacks — an example being a federal rule aimed at tightening obligations around 401(k) rollovers that lost legal traction after administrative changes. That patchwork of standards leaves consumers vulnerable to conflicts of interest even when they consult humans.

Who is on the hook when AI-informed advice harms investors?

Some legal scholars argue that current practice suggests AI companies are not fiduciaries because they do not receive direct compensation tied to personalized investment recommendations. Others point out a different risk: human advisors who use AI tools could still be liable if they rely on those tools to make recommendations that are not in a client’s best interest.

Put simply: if a registered advisor delegates analysis to an AI and the outcome harms a client, regulators and courts are likely to hold the human advisor responsible, not the platform.

Researchers say the policy response will need to evolve if AI is to become a fully trusted substitute for human advice. Until laws and regulatory frameworks clarify corporate responsibilities and consumer protections, experts expect people will continue to use AI as an informational aid — not as a drop-in replacement for fiduciary-backed counsel.