Two-Thirds of Americans Now Use GenAI for Financial Advice - What Finance Pros Should Do Next
Two-thirds of Americans are already tapping generative AI for money advice, per an Intuit Credit Karma survey released in September. A similar share say they use it often. Three in four feel more comfortable asking AI questions they'd be embarrassed to ask a human advisor.
This isn't a fad. It's a client behavior shift. The advisors who win will integrate AI into their workflows without handing over judgment.
What people ask AI most
- Financial education and basic personal finance concepts (35%)
- Financial goal setting and action plans (35%)
- Budgeting and expense management (34%)
- Optimizing savings (33%)
- Investing in the stock market (32%)
Source: Intuit Credit Karma
Is GenAI a trustworthy advisor?
Large language models can deliver surprisingly decent personal finance guidance, according to MIT economist Andrew Lo. Early analysis suggests LLMs can reach near "passing" domain knowledge with modest tuning - close, but not perfect.
Consumer confidence is high. Four in five users reported better finances after using AI, and 81% felt more confident managing money. Among those who tried it, 79% found the information accurate and 71% found it helpful. Reference: MIT Sloan.
The risk signal advisors can't ignore
It's not all upside. Over half (52%) said they made a poor financial decision based on AI's advice. The good news: 80% who acted on AI input also researched and validated it, which hints at a built-in appetite for human oversight.
Translation for practitioners: position AI as the first draft, not the final word. Your edge is context, judgment, and accountability.
How to integrate AI into your practice this quarter
- Set clear use cases: client education, budgeting drafts, goal planning outlines, portfolio explanations, meeting summaries.
- Create a verification checklist: require sources, re-calc key numbers, run scenario tests, and compare AI outputs against firm playbooks.
- Standardize prompts: ask for assumptions, trade-offs, risks, and alternatives in every output. Enforce "show your work."
- Protect data: strip PII, use enterprise controls, limit retention, and log prompts/outputs for audits.
- Compliance first: disclose AI use, document human review, and archive artifacts. Build model-use policies and exceptions.
- Client experience: let AI handle FAQs and pre-work; you handle interpretation and decisions. Make the "human-in-the-loop" obvious.
- Measure outcomes: track accuracy, time saved, client sentiment, and error rates. Prune use cases that don't clear the bar.
- Upskill the team: short, recurring training beats one-off workshops. Assign an internal AI lead to own standards and QA.
Practical prompts you can deploy
- "Draft a 90-day cash flow plan for a household with X income, Y expenses, and Z goals. List assumptions, risks, and 3 options with pros/cons."
- "Summarize this client's spending by category from the CSV. Flag anomalies and suggest 5 savings actions with expected monthly impact."
- "Explain this 60/40 proposal in plain English for a first-time investor. Include fees, tax notes, and what could go wrong."
Bottom line
Your clients are already using AI. Meet them there with structure, oversight, and clear value where AI falls short. Use the machine for speed and breadth; use your expertise for nuance and decisions.
Helpful resources
Your membership also unlocks:
