After Automation

After Automation: Why AI May Be Creating More Human Work, Not Less

There’s a strange paradox sitting right at the center of AI right now. The tools keep getting better, faster, and more capable, and yet for many teams, the amount of meaningful human work doesn’t shrink. It grows.

That’s the core idea behind Every’s After Automation, and it’s a timely one. The company has pushed automation as far as it can, using Codex, Claude Code, and a growing mix of agents across coding, writing, design, and customer service. And still, the humans haven’t disappeared. In fact, they’re busier than ever.

That may sound backward at first. If AI can write, code, answer emails, and triage support tickets, shouldn’t the workload lighten? Maybe a little. But what seems to happen instead is more interesting. AI clears away the repetitive stuff, and what’s left is the messy, judgment-heavy work that only humans can really steer. The kind of work that needs taste, context, and a steady hand. You know, the parts where someone has to actually decide what “good” looks like.

At Every, that shift is already visible. Managers are coding. Engineers are talking directly to customers. AI is handling most email replies, but humans still review them. Slack messages might come from a person or an agent, and sometimes you can’t tell right away. It’s efficient, but also a little uncanny. Familiar, but tilted slightly to one side.

The article also makes a strong point about how AI is changing the value of expertise. Once a task becomes easy to copy, default output gets cheap. What becomes valuable then is the thing AI can’t quite flatten, the unusual perspective, the sharp edit, the specific decision, the human call.

That’s probably where work is headed. Not toward empty offices, but toward more collaboration, more oversight, and more human judgment layered on top of machine speed. A little weirder. A lot more dynamic.

If you’re trying to understand what work looks like after automation, this is a thoughtful place to start.