Sarah Friar argues in OpenAI for redesigning finance work around AI

Sarah Friar argues in OpenAI for redesigning finance work around AI

Sarah Friar’s piece was published on August 10, 2026. She argues that building an AI-native finance function means redesigning work around important decisions, not merely adopting new technology or accelerating routine tasks. At OpenAI, the ambitions are a “zero-day close”—a real-time, reconciled, traceable view of the company’s finances—and continuously updated forecasting. Friar says both remain works in progress; the changes so far have moved the team toward live tools and given finance professionals more ability to build tools for their work. Her practical argument is to start with a consequential decision, connect the data and approvals it needs, and let AI analyze, coordinate, or complete parts of the workflow. People still validate numbers, exercise judgment, and own sign-off. Friar also urges CFOs to pair experimentation with clear controls and assess AI by the work completed, its cost and quality, and whether it helps people act faster or make better decisions—not by usage alone. She presents finance as a working model for broader company change: teams with more capacity and leaders with a clearer view of the business. The piece describes OpenAI’s experience but gives no measured results for the broader outcomes it advocates.

Publication: bf09c005-fdc2-3fdf-c86d-a405e793e7be

attribution — pass: The page credits Sarah Friar as author on OpenAI's site, and the reading attributes the argument to her consistently. Her own claims, such as the working-model claim, remain attributed rather than presented as established fact. This is one first-person essay, not a listing. The Keep reading links are not used.

Author Sarah Friar

summary — pass: The reading leads with Friar's thesis, which is to redesign work around decisions. It explains zero-day close in plain language, preserves the 'still building' condition and human sign-off, and summarises the scorecard. Although this is OpenAI writing about its own use, the reading neither names nor promotes products such as Codex or ChatGPT Work. The question follows from the source's own measurement framework.

Getting there requires more than adopting new technology. It requires redesigning work around the decisions that matter

date — pass: The originalPublished instant, 2026-08-10, matches the source dateline at day precision, and the reading discloses the archive copy. The Keep reading sidebar includes an Aug 11 item, which is likely a later page addition; the reading does not use or rely on it. The ambitions are reported as ongoing, not as outcomes.

August 10, 2026 Company What building an AI-native finance function taught me

support — pass: Each claim matches a verbatim source passage. The limitation that no measured results are given for the broader outcomes is fair: the source offers anecdotes, such as hours to seconds for first drafts, and research on usage shares, but no outcome metrics for the zero-day close or forecasting. The reading presents no unsupported benefits, and it keeps the stated benefits as Friar's claims.

We are still building toward both ambitions. The work has already changed how our team operates.

media — pass: No media was selected, and no verified media candidates were supplied, so there is nothing misidentified or irrelevant.