According to Gartner, the rapid adoption of AI coding agents is set to significantly increase software development costs, with AI coding expenses expected to surpass the average developer’s salary by 2028. The primary driver is the surge in large language model (LLM) token consumption, coupled with the industry’s shift from fixed seat-based licensing to consumption-based pricing models.
Gartner notes that while organizations are scaling AI-assisted development to boost productivity, many underestimate the financial impact of token usage. Developers often prioritize speed and convenience, resulting in higher token consumption and escalating costs. Limited transparency from vendors around token billing further complicates cost forecasting and budget management.
The analyst firm also highlights governance gaps as a major contributor to rising expenses. Uncontrolled agent autonomy, oversized context windows, and inadequate monitoring mechanisms can quickly inflate AI spending. At the same time, many AI coding platforms still lack mature cost-optimization features.
To address these challenges, Gartner recommends that organizations establish clear AI usage policies, align model selection with task complexity, optimize context engineering practices, implement token governance controls, and regularly review high-consumption workflows. These measures can help engineering leaders balance AI-driven productivity gains with sustainable cost management.
As AI coding tools become mainstream across software teams, organizations will need stronger governance frameworks to ensure that growing AI investments continue to deliver measurable business value.