AI Coding Costs Predicted to Surpass Developer Salaries by 2028, Report Warns

A new analysis predicts that by 2028, the cost of using AI coding tools will exceed the salaries of the developers they assist, according to experts cited in a TechRadar Pro report. The finding underscores a looming budget crisis for organizations that have rapidly adopted AI assistants without tracking the associated costs.

The report notes that most organizations currently have no system to monitor AI token consumption, leaving them blind to expenses that could soon match or exceed annual developer compensation. Without intervention, AI coding costs are projected to grow at a rate that outpaces salary increases.

The Hidden Cost of AI-Assisted Development

AI coding tools such as GitHub Copilot and ChatGPT charge per token, and heavy usage by developers can quickly accumulate significant bills. The analysis suggests that as these tools become more integrated into workflows, the per-developer cost of AI assistance could surpass the median developer salary in many markets by 2028.

• AI coding costs are projected to grow at roughly 30% annually, while developer salaries rise at about 5% per year.
• Most companies lack dashboards or alerts to track AI token spending per developer or team.
• The trend could force organizations to reallocate budgets from hiring or training to AI tool subscriptions.
• Developers often have no incentive to limit token usage, leading to unchecked consumption.

Implications for IT Budgets and Developer Productivity

The report warns that token discipline will not emerge through developer choice alone. Management must implement usage policies, set spending caps, and invest in monitoring tools to avoid cost overruns. Some organizations may need to restrict AI tool access to specific tasks or teams.

Looking ahead, the analysis calls for greater transparency from AI vendors on pricing models and for internal IT teams to build cost-tracking systems. Without these measures, the financial burden of AI coding could undermine the productivity gains these tools promise, forcing difficult tradeoffs between automation and headcount.