AI productivity paradox
The AI productivity paradox is the gap between how productive AI coding tools feel and what they measurably deliver. Developers report working faster and produce far more output, yet at the level of the whole organisation, delivery, quality and business outcomes often do not improve to match. More is produced, and less of it can be accounted for.
The paradox appears wherever AI-assisted development is measured honestly. In a controlled study, experienced developers were 19% slower with AI on familiar codebases while believing they were 24% faster (METR, 2025). Teams have shipped a 33.7% rise in task throughput alongside a 242.7% rise in incidents per pull request (Faros AI, 2026).
It matters because the instruments most organisations use, story points, tickets and velocity, measure effort, and AI has broken the link between effort and value. Seeing through the paradox requires measuring the value the AI produces rather than the volume, independently and over time.
The AI productivity paradox, in full →
Updated
