《德勤：2026向Token经济战略性转向：驾驭AI支出新格局研究报告（英文版）（28页）.pdf》由会员分享，可在线阅读，更多相关《德勤：2026向Token经济战略性转向：驾驭AI支出新格局研究报告（英文版）（28页）.pdf（28页珍藏版）》请在三个皮匠报告上搜索。

1、 The pivot to tokenomics:Navigating AIs new spend dynamics The pivot to tokenomics:Navigating AIs new spend dynamics A note from the authors:AI economics affect most organizations and the C-suite uniquely.This paper guides those familiar with AI tokens in making strategic choices.If youre just begin

2、ning your exploration of tokenomics,look for additional research soon.2 The pivot to tokenomics:Navigating AIs new spend dynamics Traditional total-cost-of-ownership frameworks miss the reality of AI Volatile workloads,new infrastructure demands,and tokens as the practical unit of cost Across indust

3、ries,Generative AI(GenAI)has become the fastest-growing line item in most corporate technology budgetsalready consuming up to half of IT spend in some firms.1 Cloud bills are rising nearly 20%year over year,driven by AI workloads.2 At the same time,geopolitical uncertainties are intensifying calls f

4、or data sovereignty and technology infrastructure independence,making many enterprises think about AI sovereignty and gaining greater control over their infrastructure.3 This is no longer a CIO operational issue;it is a CFO-and-board capital question about how to responsibly manage an investment of

5、this scale and volatility.Unlike prior technology waves governed by licenses or virtual machines,AI spend often scales in nonlinear and unpredictable ways.AI capabilities run on tokens:small chunks of datatext,image or audiothat AI systems process in training,inference,and reasoning.Every AI interac

6、tion consumes tokens,and every token carries a cost.The complexity of AIs economics hides within these tokens.Costs rise not only with user adoption but with workload design,algorithmic complexity,and infrastructure intensity.What exactly are the thresholds to move across different consumption choic