Joseph Kim, who leads OpenAI's financial services work, joined a panel at AI Engineer New York to discuss the technical and strategic decisions behind ChatGPT for Financial Services. He explained why OpenAI chose to bundle market data connectors into the product despite existing alternatives, how financial reasoning is trained into the shared model, and the importance of workflows that allow professional analysts to review and verify outputs before acting on them.
Kim addressed a central tension in AI-powered finance: building systems that are useful enough to improve workflows but trustworthy enough that experts will actually adopt them. He touched on the limits of measuring "taste" and expert judgment in AI systems, the role of code quality as a proxy for model behavior, and how financial professionals climb the analyst stack by using AI tools. OpenAI's finance team is actively hiring.
This is an AI-written summary. The original reporting belongs to @LatentSpacePod. How we write these · Spot a mistake?