I’m as bullish on AI as anyone. But there’s a pattern playing out in companies right now that I find genuinely frustrating, and this article puts it well.
Goodhart’s Law says that when a measure becomes a target, it ceases to be a good measure. And right now, AI usage is the target. Token counts, Anthropic bills, n8n workflows shared in Slack, skills written in markdown, dashboards tracking adoption. In fact, dashboards all over the place.
Nobody is asking whether any of it is producing better results. Teams are being pushed to use the tools, not to do better work. And when you optimise for usage, you get usage. People build things that look like AI solutions the same way Soviet factories produced pig iron nobody could actually use. The metric looks great. The output is junk.
The Klarna story is a good example in the article. They tried to replace Salesforce with internal AI tooling, declared victory too early, and quietly reversed course. The productivity gains that get reported in surveys are often just people finding smarter ways to make the number go up, not the work itself improving.
If AI genuinely helps, and I believe it does, then the proof is in the results. Better output, faster delivery, fewer errors, happier customers. That’s what companies should be measuring. Not whether their employees are logging enough prompts to hit a quarterly target.