Chatham Financial has integrated OpenAI’s Codex and GPT-5.6 models to streamline its capital markets advisory services, aiming to reduce time spent on routine execution while maintaining auditability. The firm is leveraging these tools to build internal applications, enhance its next-generation operating system Chatham Onyx, and reengineer consulting workflows through its Process Zero service.
What Happened
Chatham Financial is using OpenAI’s Codex to develop internal and client-facing tools, alongside GPT-5.6 to power AI features across its platforms. According to the company, this initiative supports its Process Zero consulting service, which focuses on reengineering workflows around specific outcomes. A key example is the automation of trade validation: using Codex, Chatham built an application that gathers transaction evidence, compares key terms, and flags discrepancies. The company reports that in early measurements, this application reduced review time from approximately 30 minutes to under 4 minutes. Chatham states it is currently validating these results against experienced reviewers before expanding the automation.
Internally, employees use ChatGPT and Codex for research, analysis, and software development. The firm has established Chatham Vibes, an internal platform where employees build applications tailored to their work. These apps default to GPT-5.6 Terra for AI features, with GPT-5.6 Sol available as an optional upgrade based on configuration. These tools support workflows such as reviewing maturing-cap trades, producing fixed-income rate sheets, and preparing hedging dashboards.
Chatham is also using Codex throughout the development lifecycle of Chatham Onyx, its capital markets operating system. The platform integrates assets, debt, and derivatives into a governed environment. Onyx leverages multiple models, including GPT-5.6 Sol, GPT-5.6 Terra, GPT-5.4, and GPT-4.1. Simpler tasks are routed to cost-effective models, while complex tasks requiring high accuracy use GPT-5.6. One feature, ChatFIN, uses these models to summarize historical market data patterns and link to legal documents for debt and derivative terms.
Why It Matters
The integration of AI into high-stakes financial workflows highlights a shift toward using large language models for structured, auditable tasks rather than just generative content. For developers and the industry, Chatham’s approach demonstrates a model for enterprise AI adoption where human judgment remains central. By using Codex to build custom applications and routing tasks to specific models based on complexity and cost, Chatham illustrates a pragmatic strategy for scaling AI without compromising on the accuracy and traceability required in capital markets. This case study provides a blueprint for other financial institutions looking to automate routine execution while preserving expert oversight.
The Bottom Line
Chatham Financial is actively deploying OpenAI’s Codex and GPT-5.6 models to enhance its advisory services, reporting significant time savings in trade validation and improved development speeds for its Chatham Onyx platform. The firm plans to expand these applications to additional trade types and continue refining its employee-built tools, maintaining a focus on auditable and accurate results.