AI Coding Token Costs Expected to Rival Global Average Developer Salaries by 2028

Gartner predicts that within the next two years, AI coding costs will meet or exceed the typical software engineer's monthly salary. [1] This prediction is based on a global average salary of $2,000 per month, though in regions like the US, annual salaries often reach six figures. [1] The increase in costs is driven by developers increasingly adopting generative AI and agentic tools. [1] Additionally, vendors are moving away from traditional flat per-seat SaaS models toward consumption-based licensing to balance infrastructure investments with profitability. [1]

Enterprises are scaling their deployment of AI coding agents, but many still underestimate the associated token costs. [1] Gartner senior principal analyst Nitish Tyagi noted hearing extreme examples of individual developer or business user consumption reaching $20,000 to $32,000 in a single month. [1] These figures highlight the potential impact of ungoverned token costs. [1] Currently, software engineering workload cost structures are highly variable, with limited transparency into calculation and billing. [1] Tyagi also stated that AI coding vendors have not yet provided mature, built-in cost optimization capabilities. [1]

The transition to consumption-based pricing for AI coding tools represents a fundamental shift in how IT budgets will be managed, moving away from predictable SaaS licensing toward variable, usage-driven models. The examples of extreme individual spending highlight the immediate financial risks for organizations that fail to implement governance frameworks alongside AI deployment. With vendors lacking native cost optimization features, enterprises will need to develop their own monitoring and control mechanisms to prevent unexpected budget overruns. What the sources don't address: How quickly third-party financial operations (FinOps) tools will evolve to manage and optimize these specific AI token expenditures.