The article, written by OpenAI‘s CFO, describes the journey of building an AI-native finance function at OpenAI, with the ambitions of a zero-day close and continuous forecasting—giving leaders a real-time, reconciled view of the company’s financial position and a continuously updated forecast that shows how the business is changing and which decisions could alter outcomes. The author emphasizes that the real opportunity is not just closing the books faster but seeing the business as it changes and helping leaders act sooner.

The first lesson is to give everyone access to AI tools and create a reason to use them, such as through a finance hackathon that produced IR-GPT, a custom GPT for investor relations, and other GPTs for procurement and tax. The second lesson is to redesign the full workflow around the decision, mapping the path from source data to decision and letting AI handle analysis, coordination, and completion while finance validates and owns the final sign-off. The third lesson is that finance professionals become builders: using tools like Codex and ChatGPT Work, team members with no coding background can build custom dashboards and tools that turn static models into live, interactive views.

The fourth lesson is to pair speed with clear accountability and controls, ensuring that every AI output connects to a reliable source, every forecast carries a clear explanation, and every change to an approved baseline requires finance authorization. The fifth lesson is to measure value per unit of intelligence, asking whether AI completed work that mattered, what it cost, whether the result was good enough, and whether it helped move faster or make a better decision. The author concludes that the CFO has a unique opportunity to lead the company’s AI transformation from the center of strategy, capital, data, risk, and performance, creating a finance team with greater capacity and a clearer view of the business.