AI usage metrics can distort real developer productivity
means that once a number becomes the goal, it often stops being a useful measure. A company can try to measure AI adoption through , commit count, size, or an innovation percentage.
Those numbers may not show whether work is actually getting better. After one or two weeks of confusion, employees may simply change their behavior to fit the metrics.
They might make smaller commits more often and use Opus 4.8 on a high-effort setting by default. The can look busier while real remains unclear.
Key points
- warns that a measure can become less useful once people are judged by it.
- AI adoption can be distorted if it is measured only by , commit count, size, or innovation percentage.
- People may adapt to the metric instead of improving the work itself.
- Smaller, more frequent commits and default use of a high-effort model can make activity look higher.
- For AI tools, real outcomes matter more than raw usage numbers.