FinSafetyBench: Evaluating LLM Safety in Real-World Financial Scenarios
For financial institutions deploying LLMs, this benchmark identifies specific compliance risks and attack vectors that current safeguards fail to address.
FinSafetyBench evaluates LLM safety in financial scenarios, revealing critical vulnerabilities where adversarial prompts bypass compliance safeguards, with stronger susceptibility in Chinese contexts and limitations of prompt-level defenses.
Large language models (LLMs) are increasingly applied in financial scenarios. However, they may produce harmful outputs, including facilitating illegal activities or unethical behavior, posing serious compliance risks. To systematically evaluate LLM safety in finance, we propose FinSafetyBench, a bilingual (English-Chinese) red-teaming benchmark designed to test an LLM's refusal of requests that violate financial compliance. Grounded in real-world financial crime cases and ethics standards, the benchmark comprises 14 subcategories spanning financial crimes and ethical violations. Through extensive experiments on general-purpose and finance-specialized LLMs under three representative attack settings, we identify critical vulnerabilities that allow adversarial prompts to bypass compliance safeguards. Further analysis reveals stronger susceptibility in Chinese contexts and highlights the limitations of prompt-level defenses against sophisticated or implicit manipulation strategies.