GitHub, Website And App Data Shows How AI Has Led To An Explosion In Coding Productivity

The impact of AI is already showing up in the data relating to coding productivity.

According to recent analysis by FT, year-on-year growth across multiple indicators of coding output has surged dramatically since late 2024. New websites saw growth accelerate to approximately 40% by early 2026, while new iOS apps experienced an even more striking spike, reaching roughly 50% growth over the same period.

The trend is particularly evident in GitHub data from both the United States and United Kingdom. GitHub code pushes in the US jumped to about 35% year-on-year growth in 2026, while UK figures show around 30% growth. Both metrics had been relatively flat or modest throughout 2022 and 2023 before the sharp uptick.

The timing of this productivity explosion aligns closely with the widespread adoption of AI coding assistants and large language models trained specifically for software development. Tools like Cursor, Claude Code, Codex and Google Antigravity have become increasingly sophisticated and prevalent in developer workflows over the past year.

This represents a fundamental shift in how software is created. AI assistants can now generate boilerplate code, suggest completions, debug errors, and even architect entire features based on natural language descriptions. What once took hours can now be accomplished in minutes, enabling individual developers and small teams to ship products at a pace previously reserved for much larger organizations.

The data suggests we’re witnessing the early stages of a significant productivity revolution in software development. As these AI tools continue to improve and developers become more adept at leveraging them, the gap between pre-AI and AI-assisted coding productivity is likely to widen further.

For businesses, this creates both opportunities and imperatives. Companies that effectively integrate AI into their development processes stand to gain substantial competitive advantages through faster iteration cycles and reduced time-to-market. Those that lag in adoption risk falling behind as the new productivity baseline continues to rise.

Posted in AI