AI Value Is Established. Now the Focus Is on Economics
The conversation around AI is shifting. The question is no longer simply whether AI creates value, but what it costs, where organizations should invest, and how those investments should be prioritized.
As AI adoption grows, organizations need greater visibility into current and future spending. They also need the financial discipline to connect that spending to business value.
AI Costs Are More Dynamic Than Traditional IT Costs
Traditional technology investments were comparatively predictable. AI is different because its economics are heavily driven by consumption.
Token-based pricing is one reason. As users interact with large language models, both input and output tokens generate costs. Longer conversations can compound consumption because the model repeatedly processes previous context. This makes AI usage, and therefore cost, difficult to predict using traditional assumptions.
But tokens are only part of the picture. Infrastructure, security, governance, people, applications, and the surrounding technology environment all contribute to the total cost of ownership. Focusing on model costs alone can therefore give organizations an incomplete view of what AI actually costs.
The Risk Runs in Both Directions
Poor financial management can lead organizations to spend heavily on AI without a clear return. Costs rise, budgets are exceeded, and it becomes difficult to explain what the business is getting in return.
But investing too little may be the greater long-term risk. Organizations that fail to adopt AI and use it to improve their processes risk falling behind competitors that do. The challenge is therefore not simply to control spending. It is to make the right investments and understand their impact.
Use Existing Disciplines Rather Than Creating Another Silo
AI economics does not require an entirely new management practice.
FinOps, TBM, IT Financial Management, finance, architecture, security, and business teams already bring many of the capabilities required. The financial principles remain the same: understand the cost of a unit, how many units are consumed, who consumes them, and how that consumption translates into business value.
The priority is to bring these disciplines together. Creating another dedicated silo adds complexity where collaboration is needed instead.
CIOs and CFOs Need a Shared View of AI Investment
For CIOs, the immediate task is to give senior management transparency into current AI investments, run costs, and how those costs are likely to develop. That view should also include the expected value of AI initiatives.
Finance has an active role to play. Rather than simply asking IT to explain rising costs, CFOs and IT finance leaders should bring their expertise into the discussion and work with technology teams to determine where to invest and what return to expect.
Keep the Business Case Close to the Reality of AI
The fundamentals of the business case have not changed: define the cost and define the expected benefit. What has changed is how difficult the cost side is to predict.
For that reason, long-range three- or five-year business cases may be impractical for AI. Dr. Becker recommends working with a shorter horizon, potentially around 12 months, and being specific about the expected outcome of each use case. That could mean revenue improvement, customer satisfaction, productivity, or labor-cost optimization.
Get Ahead of the Problem
The central message is straightforward: organizations need to address AI economics before cost overruns force the conversation.
That means building transparency around AI spending, understanding total cost rather than tokens alone, and putting business and IT in a position to make informed investment decisions. The goal is not simply to spend less. It is to know where to invest, what it will cost, and what the organization expects to get in return.
