Following the Algorithm: How Frequency Bias Shapes AI Coding Choices

As AI coding assistants like Claude Code and Github Copilot become integral to software development, they increasingly recommend languages and tools based on their prevalence in training data rather than their suitability for the task. This frequency bias creates a vicious cycle where developers—especially newcomers—are steered toward mainstream technologies, accumulating technical debt and preventing emerging tools from gaining adoption. The research highlights a critical fairness gap in AI-assisted development and calls for greater transparency in how these tools influence programming decisions.