EY refers to the global organization, and may refer to one or more, of the member firms of Ernst & Young Global Limited, each of which is a separate legal entity. Ernst & Young Global Limited, a UK company limited by guarantee, does not provide services to clients.
How EY can help
-
Experience AI innovation at the EY.ai Lab! Collaborate with multidisciplinary teams to develop tailored solutions that address your unique business needs.
Read more
The issue is becoming more important as token pricing is ultimately tied to a constrained physical supply chain: chips, power, data centers, cloud capacity and model provider infrastructure. Enterprises may experience that constraint in the form of token bills, usage caps, rate limits, model access restrictions or unexpected repricing. For many companies pushing AI adoption, the last few months have been a shock to the budget.
Current pricing may also understate the long-term economics of agentic AI if some portion of compute cost is being absorbed, subsidized or strategically priced by upstream providers. As agentic use cases become more valuable and more compute-intensive, those costs are unlikely to remain invisible indefinitely. Someone will ultimately pay for the compute that agents consume. Many providers are already shifting from subscription-based fees to consumption-based models, ending the days of vendor-subsidized pricing.
But while tokens are the story today, the real economics of running agentic workflows at scale include infrastructure, operations, people, risk and the hidden cost of engineering around AI’s limitations. Optimizing tokens without understanding total cost of ownership is like managing a factory by watching the electricity bill.
The story previously presented to many boards was straightforward: an expensive predictable line of human labor traded for a cheaper predictable line of software. But an agent is not a license. It lives in an operating stack of compute, models, data pipelines, governance, oversight and redesign work, each of which compounds rapidly. As a result, the tokens AI consumes, serving as the metered proxy for much of that operating stack, become a highly variable operating cost.
Consider a customer service AI assistant built two years ago to answer product questions. A chat that once cost $0.04 may now involve tool retrieval, planning and subagents that turn it into a $1.20 orchestration. Additional costs — such as knowledge-base updates, agent evaluation and human-collaboration design — may not appear on the model vendor’s invoice but still surface to the enterprise.