Opinion
The next AI revolution isn't about models. It's about management
As AI becomes embedded in the enterprise, the basic unit of work is shifting. Organizations that manage this change only through cost controls may save on the wrong resource, and miss the opportunity to turn intelligence into measurable business value.
Some revolutions announce themselves loudly. Others unfold almost quietly, until it becomes clear that they have changed not only technology, but the rules by which organizations create value. The AI revolution belongs to the second kind. Much of the discussion around AI still focuses on the models themselves or on productivity gains. These are important questions, but the more meaningful shift is taking place deeper within the enterprise: in its economic model.
For more than a century, modern management has relied on three variables: people, time and money. Budgets, workplans, compensation models and forecasts were built around them. Yet this vocabulary is no longer sufficient to explain how organizations generate value.
When an AI agent writes code, reviews a contract, prepares a presentation or summarizes thousands of documents in minutes, the organization is not merely improving productivity. It is beginning to rely on a new unit of production — one that cannot be measured only in human working hours.
Tokens: The New Unit of Work
To many, the term “tokens” sounds like a technical detail. In practice, it may become one of the most important economic concepts of the coming years. Tokens are not merely the billing unit of language models; they are the consumption unit of the new workforce. Just as the work hour has long been used to measure human labor, tokens are becoming a way to measure work performed by AI models.
Many executives compare this shift to the move to cloud computing: pricing models changed, and the discipline of FinOps emerged. But the comparison misses the essence of the transformation. Cloud changed the way organizations consume computing infrastructure; it did not change the basic unit of production. Even after the move to cloud, people remained the creators of value. AI changes that assumption. For the first time, organizations are relying on a non-human productive force. This means that not only the technology is changing — the economics on which the organization relies are changing as well.
If we once measured employees and hours, we will now need to measure models and the intelligence they consume. The central question will not be only how much AI costs, but what return is generated by each unit of consumed intelligence.
Fighting the Wrong War
This is where a common managerial mistake may emerge. As AI budgets grow, many organizations will respond with usage limits, quotas and pre-approvals. Any reduction in token consumption will be presented as a success.
Yet this may be precisely the strategy that harms competitiveness. A factory does not succeed because it saved steel, but because it produced more value from every ton it purchased. In the same way, an AI-based organization will not win because it consumed fewer tokens, but because each token generated more business value.
The impact will not stop at the technology budget. It will also affect how organizations measure employees, reward them and build teams. Many performance models still rely on time, workload or the number of tasks completed. Yet an employee who uses AI tools well can already produce output that once required an entire team.
This does not mean that human talent is becoming less important. On the contrary: as more tasks are performed with AI, the value of people who can exercise judgment, ask the right questions, review outputs and translate technological capability into business outcomes will only increase.
That is why the AI discussion cannot remain solely in the hands of technology departments. The CIO will need to explain how models are selected and how the value they generate is measured; the CFO will need to understand the economics of intelligence; HR will need to update performance models; and control, risk management and corporate governance functions will need to redefine their success metrics.
In other words, token economics is no longer just the responsibility of the CIO. It is becoming a new management language for the organization.
A decade from now, it may be hard to understand how organizations tried to manage AI-based enterprises using models built for a world in which only humans produced work. Just as it is difficult to imagine a cloud environment without FinOps, it may become difficult to imagine an AI-based organization without a discipline that measures and manages the intelligence it consumes.
Organizations that understand this first will not manage AI as another expense line, system or innovation project. They will build a management language that measures not only work, but capability; not only cost, but return; and not only employees, but the combination of people, machines and intellectual capital. Those that fail to adopt this mindset will not fall behind because their model is slightly less intelligent. They will fall behind because they continue to manage a 21st-century organization with an economic model built for the 20th century.
Noam Gonen, Partner, Chief Strategy Officer, Deloitte Israel; Inbal Namir, AI Workforce Transformation Leader, Deloitte Israel; Adam Zeitlin, FinOps Lead, Deloitte Israel.
