CFO Sarah Friar introduced a new framework to quantify AI value through four metrics: useful work, cost per successful task, dependability, and return on compute. This shift moves the conversation from vague productivity gains to hard financial data. Practitioners can now benchmark OpenAI deployments against concrete cost-to-output ratios to justify enterprise spend.