Productivity is often conflated with value. You take raw inputs and transform it into something more valuable. Trees into paper, steel into skyscapers, ideas into blogs (maybe not that).

Value is even more complicated to calculate. It is the difference between the price of a coffee bean ( pre-tariff of course) and a cup of coffee at a high end cafe at some high rent location as one my favorite speakers, Joe Pine, wrote in The Experience Economy many years ago. The value of a particular good is very much tied to an associated experience. That is more than just inputs and outputs.

It is bundling ideas with goods to create something new.

AI productivity is a slippery slope. Generating more derivative content in less time doesn’t deliver more value. Personally, it feels like you struck gold. Write a clever prompt and you get back lots of information. That sure feels rewarding and personally productive, but what did you really create. On your first read, the response will seem thorough and satisfying, but it won’t have created any particularly fresh or valuable insight. If you think “corporate speak” is bad now, just wait until all conversations are filtered through the voice of AI

Now, if you are writing code for a project, it can save you time. But it is the total output of the end product that ultimately reflects its value, not just your code.

Measuring the real productive output of AI is a little trickier than just how much time did you save. The coding, the infrastructure, the energy and the training data needed are not part of your personal productivity metrics. Maybe you count your subscription fee and the effort of making your prompts part of the productivity equation. That works fine on a personal level, but if AI is going scale as an enterprise system, it is going to need to know to interact collaboratively.

AI is going to get there at some point, but I don’t necessarily see it as an enterprise system productivity tool yet. Individuals using AI to help with their tasks may feel like it is productive, but does it really create new value. Producing notes from a meeting that probably didn’t need to happen doesn’t qualify as corporate productivity. It is a real time saver for the designated note taker of the meeting, but after a while these meeting notes start to pile up redundantly in people’s archive folders. Meanwhile, a meeting with no AI assistance that surfaced sticky issues and made some difficult decisions that the team rallied enthusiastically behind would be far more productive.

The value of an enterprise is in the coordination of resources to deliver goods or services that customers value. It is not necessarily in speeding up the work of individual contributors or increasing their ability to generate more information. Just as in basketball, sometimes it is necessary for a player to miss a free throw in order to give the team a chance to score the points that it really needs. AI in the enterprise needs to be about helping organizations achieve more than the sum of the individual players. Thinking about how AI can be used to speed up the output of the enterprise is a more productive use of your time.

Next week we will look at the absurdity of enterprise apps in general and the unimaginative ways that they are adding AI capabilities.