What Are We Actually Going to Do With the Time?

I just finished a project in three days that would have taken three weeks two years ago. Same deliverable. Same quality bar. Dramatically different timeline. And when I was done, I sat with an uncomfortable question I hadn't expected to be asking: what am I supposed to do with the rest of the week?

That question sounds like a good problem to have. It is. But it also points to something we haven't figured out yet as an industry, or as a society. AI is compressing the time it takes to produce knowledge work in ways that are real and measurable. And we have no framework for what that means.

We Inherited a Schedule Designed for Factories

The 40-hour work week was a hard-won response to the Industrial Revolution. Factory workers in the 19th century routinely worked 60 to 80 hours in dangerous conditions. The labor movement fought for decades to establish a humane standard. Henry Ford popularized the 40-hour week at his factories in the 1920s. The Fair Labor Standards Act codified it in 1938.

That standard made sense for physical labor at fixed machinery. It made less sense when the economy shifted to knowledge work in the second half of the 20th century, but we kept it anyway. We designed office buildings, commuting infrastructure, childcare systems, and social expectations around a schedule that was optimized for steel production in 1926.

Nobody ever really asked whether it fit the new kind of work. We just inherited it.

Now the work is changing again, faster than before. And the inherited schedule is going to come under pressure in a way it hasn't since the labor movement fought for it in the first place.

The Multiplier Is Real

I want to be precise about what I'm describing, because there's a lot of noise around AI productivity claims. I'm not talking about using AI to autocomplete sentences or generate first drafts of emails. I'm talking about using AI as a genuine collaborator on complex, production-grade technical work.

Over the past year I've built federal software platforms, migrated legacy enterprise systems, produced government RFP responses, and designed integration architectures — with AI as an active participant in every phase. The methodology I developed across those projects is documented in a book I recently finished. The honest summary of what I found is this: the multiplier is real, it's significant, and it doesn't come from AI doing the thinking. It comes from AI removing the friction between thinking and building.

That distinction matters for what comes next. Because if AI were replacing the thinking, the value would flow to whoever owns the AI. But if AI is removing friction while the thinking remains human, then the value flows to whoever has the most experience to direct it. That's a very different economic outcome — and a much more interesting one for people who've spent decades building expertise.

Two Ways This Goes

When a genuine productivity multiplier arrives in an industry, it tends to go one of two ways.

The first way: the gains get captured as margin. One person does the work of three. Two people lose their jobs. The remaining person works the same hours for the same pay while the company's earnings improve. Shareholders benefit. Nobody gets their time back.

The second way: the gains get redistributed as time. The work week compresses. The same output gets produced in fewer hours. People get evenings back, weekends back, years of commuting back. Productivity improves not because people are working harder but because the tools got better.

Historically, the first way is more common in the short term. The second way tends to win in the long run, but only after significant friction — labor organizing, regulation, cultural shift. The gap between the two can span decades.

I don't think we're going to wait decades this time. The compression is happening too fast, and the people experiencing it firsthand are not factory workers with limited negotiating leverage. They're knowledge workers with options. The remote work shift showed what happens when that group gets a taste of something better. They don't easily give it back.

What I'm Actually Doing With the Time

I started my business as a remote operation fifteen years ago, before that was normal. The structure gave me something I didn't fully appreciate at the time: presence. I was home. I watched my kids grow up in a way most of my peers, who were spending two hours a day in traffic and missing dinner most nights, didn't get to.

AI is offering a second version of that gift to anyone willing to take it. Not just remote flexibility, but genuine time compression. The ability to do serious, high-quality work in fewer hours and use what's left for something other than more work.

I'm choosing to take it. That means being deliberate about which projects I take on, how I structure my days, and what I'm protecting time for. It doesn't mean working less — it means working in a way that doesn't require the 60-hour week to produce results that used to require 60 hours.

Most organizations haven't figured out how to receive this yet. The instinct is still to fill the reclaimed hours with more tasks. That's the path of least resistance. It's also the one that leaves the actual gain on the table.

The Question Worth Sitting With

I'm not suggesting everyone can or should immediately restructure their work life around AI-enabled time compression. The structural barriers are real — employment contracts, billing models, client expectations, organizational culture. These don't change overnight.

But the question is worth sitting with, especially for people who are already experiencing the multiplier. When the work that used to take a week takes two days, what do you do with the other three days? What would you want to do with them? What would you want your team to do with them?

Those aren't rhetorical questions. They're the practical design problem that comes after the productivity gain. And the people who answer them thoughtfully — rather than defaulting to "fill the time with more deliverables" — are the ones who will actually benefit from what AI makes possible.

The tools changed. The schedule hasn't. That gap is where the opportunity lives.

These themes are explored further in the final chapter of The Architect and The Navigator, now complete on Leanpub. Eight chapters covering the methodology, the case studies, and where AI-accelerated development is headed.