Why AI Bills Are Confusing the C-Suite: The Shift to Usage-Based Pricing Explained

A recent KPMG survey reveals that nearly a third of corporate leaders struggle to understand and control operating costs when scaling enterprise AI deployments. As AI vendors like Anthropic, OpenAI, and GitHub move from flat-rate subscriptions to usage-based pricing, organizations are finding it challenging to forecast, monitor, and manage AI spending effectively.

Key Findings

• 29% of senior leaders have difficulty understanding their AI operating costs.
• 33% cite limited understanding of AI costs and economics as a barrier to deploying AI agents.
• Nearly half of organizations have rephased AI deployments when costs outweighed expected value.
• Lower-cost, high-fidelity models are the fastest-growing influence on AI strategy, up 7 percentage points from Q1.

The Bigger Picture

Amazon and Microsoft are ramping up capital expenditure—Amazon plans $200 billion this year, Microsoft $190 billion—to build AI capacity. Both are investing heavily in forward-deployed engineering to help customers develop AI applications. Amazon has allocated $1 billion for AWS Forward Deployed Engineering, while Microsoft is providing $2.5 billion for its new Microsoft Frontier Company.

Governance Challenges

Beyond costs, AI governance remains a concern. KPMG emphasizes that executive accountability is crucial, but governance succeeds through day-to-day practices. Organizations need clear rules for human intervention, cost ownership, output review, and system failures. However, most organizations have not fully embedded these mechanisms.

Interestingly, KPMG itself faced scrutiny when GPTZero found that only 5 of 45 citations in its October 2025 report were accurate, highlighting the risks of AI-generated content without proper oversight.

What This Means for Marketers

For marketing professionals, understanding AI pricing models is essential to budget planning and ROI assessment. As AI tools become integral to marketing strategies, managing costs and ensuring governance will be key to successful adoption.