SuperMoney’s Sense AI Hits 99.4% Accuracy on Financial Planning Test

The Santa Ana-based firm tested its agentic finance app using a self-authored question bank that spans all eight official CFP Board knowledge domains. To ensure validity, SuperMoney conducted seven consecutive runs on its live production system, isolating each session to prevent the model from leveraging data from previous queries. The resulting data showed high consistency, with a standard deviation of only 0.11 percentage points.

While the Insurance, Retirement, and Psychology of Financial Planning domains achieved perfect scores across every trial, the model’s performance remained robust even in its lowest-scoring category, Tax Planning, at 97.97%. CEO Miron Lulic noted that the project was a response to the risks of AI hallucination in fiscal management, emphasizing that companies making claims about financial competence must be prepared to provide empirical evidence of their accuracy.

Because the actual CFP certification exam materials are proprietary, SuperMoney developed its own testing framework based on the Board’s publicly available 2026 blueprints. The company has released its full methodology alongside the report, though it clarifies that Sense AI serves as an educational tool rather than a substitute for professional legal or tax counsel. The assistant is currently active within the SuperMoney app.