How CFOs Can Tell If AI Is Creating Value
· news
The AI Value Gap: A Test of Corporate Willpower
A recent survey by Bain & Company reveals a stark disconnect between CFOs’ expectations and actual returns on investment in artificial intelligence. Despite increasing spending, 69% of finance leaders are either uncertain or underwhelmed by their AI investments.
CFO Sarah Friar of OpenAI is challenging the status quo with her innovative metric: “useful intelligence per dollar.” This approach focuses on tangible results rather than vendor promises, offering a much-needed shift in how companies measure success in AI deployments. By prioritizing useful outcomes over model specifications or cost per token, finance teams can make more informed decisions about their AI investments.
The numbers are telling: 56% of senior finance executives plan to increase enterprise-wide AI investment by more than 15% this year, while 83% expect budget hikes above 15% over the next two years. Yet, only 31% of CFOs rate AI outcomes in finance as strongly positive. This paradox highlights the widening gap between those who have successfully scaled AI and those who haven’t.
The AI Scorecard: A Necessary Framework
Friar’s scorecard for AI investments is a call to action that pushes leaders to prove whether AI-completed work compounds faster than its production cost, or risk reallocation before hype turns into drag. By asking four fundamental questions – what truly matters in our business, how do we quantify costs, can employee outputs be relied upon, and does AI deliver value beyond its initial investment? – finance teams can redefine their approach to AI and focus on tangible outcomes.
This scorecard reflects the industry’s growing recognition of AI value creation as a strategic imperative. As more companies take AI seriously, Friar’s metric will become a benchmark for measuring success. Will other CFOs and finance leaders follow suit, or will they continue to rely on vendor promises and speculative projections?
The Burden of Proof
The burden of proof now rests with corporate leaders who claim that AI is a key driver of their growth strategy. It’s no longer sufficient to merely invest in AI without concrete evidence of its value creation potential. As Friar suggests, “useful intelligence per dollar” should be the new north star for finance teams navigating this complex landscape.
This shift in focus raises important questions about the role of AI in driving business value. Can companies truly rely on AI to automate tasks, improve efficiency, and enhance decision-making? Or will they continue to struggle with integrating AI solutions into their existing infrastructure?
Scaling AI: A Wider Pattern
The story of AI adoption is not unique to the finance industry. Across various sectors, companies are grappling with the challenges of scaling AI deployments. From healthcare to manufacturing, early promises of efficiency and innovation often give way to disappointment and disillusionment.
Friar’s scorecard is a timely reminder that value creation in AI requires a more nuanced approach. By focusing on tangible outcomes rather than vendor promises, companies can begin to close the gap between those who have successfully scaled AI and those who haven’t.
The Next Chapter
As corporate leaders move forward, they must prove that their AI investments are paying off. Friar’s scorecard will become a benchmark for measuring success in AI value creation. Companies like OpenAI, which have already demonstrated significant growth through AI-driven innovation, will need to scale their deployments without losing sight of the underlying business drivers.
By prioritizing “useful intelligence per dollar,” companies can create a sustainable framework for AI adoption that benefits both the bottom line and employees. Ultimately, Friar’s scorecard is a clarion call to corporate leaders who must now confront the reality of their AI investments. Will they rise to the challenge, or will they continue to struggle with integrating AI solutions into their existing infrastructure? Only time will tell.
Reader Views
- ADAnalyst D. Park · policy analyst
The AI scorecard proposed by CFO Sarah Friar is a much-needed attempt to bridge the gap between lofty expectations and lackluster returns on investment in artificial intelligence. However, it's crucial to recognize that every industry has its unique value chains, and what constitutes "useful intelligence per dollar" can vary significantly across sectors. As such, finance teams should not only focus on developing a generic scorecard but also tailor their metrics to suit the specific needs of their business.
- CSCorrespondent S. Tan · field correspondent
While CFO Sarah Friar's metric of "useful intelligence per dollar" is a step in the right direction, it's essential to acknowledge that AI value creation often depends on the industry and specific use case. The scorecard approach may not be applicable across all sectors, particularly those with rapidly evolving technologies or highly variable costs. A more nuanced understanding of how different businesses can adapt Friar's framework to their unique needs is necessary to unlock its full potential and avoid forced standardization that could hinder innovation.
- CMColumnist M. Reid · opinion columnist
While CFO Sarah Friar's scorecard for AI investments is a step in the right direction, companies should also consider a critical but often overlooked aspect: data quality. Simply put, if your AI system is fed garbage data, don't expect gold to come out of it. Companies need to take ownership of their data infrastructure and ensure that the inputs feeding into these systems are accurate, complete, and reliable. Without this foundation, all the "useful intelligence per dollar" in the world won't salvage a poorly designed AI system.