The AI Productivity Trap: More Output, Less Focus

AI gave us output. It took our focus.

Doing more because we can

Time has become a currency, which we “spend” instead of “pass.” — Juliet Schor

AI was going to do the work we didn’t want to do, so we could focus on what mattered. What do we want exactly? Apparently, more. AI increases the amount of work that can be done, which in turn creates more work we can do with AI. We optimize and increase the volume, not because anyone forces us to, but because we can, and why not?

And with agents, AI given a task and the autonomy to complete it, we can run multiple workstreams at once. More work means more context, and context switching between multiple projects. We’re not getting less busy; we’re just becoming more capable and efficient at the work we produce.

The Cost

If we’re running multiple tasks and agents, we have to maintain context and understanding across them. This is mentally exhausting as we’re dividing our attention and switching focus so often. Our brains are not meant to multitask. We’re overstimulated and unfocused. We’re taking in too much information to process, and we can’t really focus on anything. Technology has done this to us before: radio, TV, and phones. Every new medium attracts more attention and shapes how we think.

Every notification, every new tab, every feed refresh trains the brain to crave the next thing rather than staying with the current one. The depth goes away. We skim instead of reading. We react instead of thinking. We start things rather than finish them. We prefer overviews rather than having to digest the information ourselves. And the more we consume without intention, the more that consumption shapes how we think, the same way social algorithms decide what we see, until we stop and set boundaries. The sheer amount of consumption distracts us from a focus point.

And AI is introducing the same pattern. More agents, more tasks, more outputs to review, more consumption. We’re not focusing on one thing and doing it well. We’re managing a volume of generated content that keeps us busy but scattered, always monitoring, never deep in anything. The productivity looks real, but the focus point is gone.

The Alternative

If we build a system (a kitchen), define a singular set of tasks (a recipe), and follow that plan to completion (cooking), we can manage our focus. There’s a start, a sequence, and a finish. The focus point is wherever we are in that process.

Not everything needs our hands on it. Food prep exists for a reason; it’s the bounded, predictable work you do ahead of time so you can focus on the actual cooking. AI agents should work the same way. Give agents tasks they can complete with autonomy, and keep your attention on the work that needs it. We can’t do everything, even with AI. And we don’t need to. We don’t need to over-consume or be wasteful, making more food than anyone can eat.