Evaluating AI for Finance: Is AI Credible at Assessing Investment Risk?
This highlights a critical credibility and fairness problem for financial institutions and regulators, as AI systems fail to maintain consistency across demographics, making it incremental by exposing flaws in existing methods.
The study assessed whether AI systems can credibly evaluate investment risk appetite by testing proprietary and open-weight models on curated user profiles, finding that models like GPT-4o assign higher risk scores based on irrelevant attributes like country or gender, violating regulations.
We assess whether AI systems can credibly evaluate investment risk appetite-a task that must be thoroughly validated before automation. Our analysis was conducted on proprietary systems (GPT, Claude, Gemini) and open-weight models (LLaMA, DeepSeek, Mistral), using carefully curated user profiles that reflect real users with varying attributes such as country and gender. As a result, the models exhibit significant variance in score distributions when user attributes-such as country or gender-that should not influence risk computation are changed. For example, GPT-4o assigns higher risk scores to Nigerian and Indonesian profiles. While some models align closely with expected scores in the Low- and Mid-risk ranges, none maintain consistent scores across regions and demographics, thereby violating AI and finance regulations.