CFO Sarah Friar introduced a new framework to measure AI return on investment. The scorecard tracks useful work, cost per successful task, dependability, and return on compute. It moves beyond vague productivity metrics to quantify actual output. Practitioners can now benchmark OpenAI deployments using concrete financial and operational data to justify compute spend.