Gartner analysts warn that enterprise spending on generative AI token usage could match or exceed the monthly salaries of software engineers within the next two years due to consumption-based pricing.

Key Points

• Gartner predicts AI token costs will rise significantly as businesses adopt agentic tools and consumption-based licensing models.
• Senior principal analyst Nitish Tyagi reports instances of individual developers and business users consuming up to $32,000 in monthly token fees.
• Current AI coding platforms lack mature, built-in cost optimization features, making it difficult for enterprises to forecast or control expenditures.
• Rapidly evolving agent-driven workflows and bloated context windows often lead to unexpected budget depletion and challenges in calculating return on investment.

Why it Matters

Unchecked token consumption threatens to erode the productivity gains promised by generative AI tools if organizations fail to implement strict governance frameworks. Businesses must prioritize cost management strategies now to prevent infrastructure expenses from outpacing the actual value delivered by their engineering teams.