A scorecard for the AI age

The artificial intelligence industry has reached a critical inflection point where organizations must move beyond hype and establish concrete methods for measuring AI's actual business impact. Sarah Friar, Chief Financial Officer of OpenAI, has introduced a comprehensive AI scorecard designed to help enterprises quantify return on investment and justify continued spending on artificial intelligence initiatives. This framework addresses a persistent challenge facing corporate leaders: determining whether their AI investments are genuinely delivering value or merely consuming resources.

Friar's proposed scorecard centers on four fundamental metrics that provide measurable indicators of AI system performance. The first pillar, "useful work," focuses on tangible outputs that directly contribute to business objectives. The second metric, "cost per successful task," measures economic efficiency by calculating expenses associated with completing specific functions. "Dependability" serves as the third component, ensuring that AI systems perform consistently and reliably across repeated applications. Finally, "return on compute" examines the productivity gains relative to computational resources invested, creating accountability for infrastructure spending.

• Organizations can now move beyond subjective assessments to data-driven AI ROI evaluation
• The framework enables meaningful comparison of different AI implementations across departments and companies
• Companies can identify underperforming AI projects and redirect resources more strategically
• Cost accountability becomes clearer, addressing CFO concerns about runaway AI budgets
• The metrics provide transparency for stakeholder reporting and investment decisions

As AI adoption accelerates across enterprises, the gap between AI spending and demonstrable returns has widened considerably. Friar's scorecard fills a critical void by providing a standardized approach to evaluation that resonates with financial decision-makers. Rather than relying on technical benchmarks or vendor claims, this framework emphasizes practical business outcomes. For CFOs, boards, and technology leaders, these metrics offer a pathway to responsible AI investment—distinguishing genuine productivity gains from technology theater. As enterprises mature in their AI journey, this disciplined approach to measurement will increasingly separate successful implementations from cautionary tales.