This page collects 3 real-world use cases of AI agents in financial services. 1 is verified by a production case study or reproducible demo; 2 are unverified reports or concepts, and they are labeled as such. The most common autonomy level here is Level 3 (Conditional): autonomous within guardrails; humans handle exceptions. Business functions covered include Accounting/Finance and Customer Service. Vendors and tools mentioned include Coinbase, Pyth Network and OpenAI. Each entry lists the workflow, tools, human-in-the-loop points and risks, and shows outcomes only when a source reports them, with a link to that source.
Coinbase CEO says a Qwen3.5-9B model fine-tuned on internal data outperformed frontier models on fraud prevention at lower cost and latency.
Exa is piloting Pyth real-time/historical market data so research agents can pair news with prices in one flow.
OpenAI-powered assistant handles refunds, returns, disputes and payment queries in 35+ languages (reported by Klarna)