Why Off-the-Shelf AI Doesn't Understand Money — Udi Menkes, Intuit
Off-the-shelf LLMs flip financial recommendations with minor assumption changes, making them untrustworthy for money advice.
“the advice sounds good. It sounds sound, but I can't really trust it.”
Intuit's Udi Menkes demonstrated that frontier LLMs produce contradictory financial recommendations when inputs change slightly, using his own real estate vs. stock market decision as a live example. Intuit is conducting a study across 100,000+ small business situations to build outcome-grounded financial models as an alternative. The talk frames generic LLMs as fundamentally unfit for financial advice without domain-specific grounding on real outcomes.