Through most of history and prehistory, daily existence required sweating. Humans mainly got to their fields or their shops on their own two legs. Over time we incorporated animals, then motors and big trains to move goods and each other, further and faster. More recently, we developed the technology to see and talk to someone anywhere on the planet at near instant time. Companies exist now without anyone ever meeting face-to-face. A rumor can start wherever and spread to half the world in a matter of hours. We have solved the problem of distance.
Except as we have increased the speed of seemingly everything we could need, one speed has not changed in all of human history - the speed of our legs. I'm not talking about Usain Bolt's 100m record but rather our quotidian walking for sustenance. Even as our society has adapted to cars and the internet, our neighborhoods still have constraints related to the distance we are willing to walk. Sure, people drive all the time to reach great distances, but people don't live in cars. Plus, cars come with traffic and parking stress, not to mention their heavy carbon footprint. Though our society has evolved dramatically to incorporate new technologies, we cannot evolve away from the necessities of walking.
Moreover, even as transportation has changed, many aspects of life remained surprisingly constant. An Italian physicist noted that average commute time has stayed around 1 hour/day, now immortalized as Marchetti's Constant. Cars didn't give us time back, they just allowed people to live further away. If anything, the commute got worse, with more stress and fewer human interactions. It is only with well-designed public transit that we are best able to leverage the advances in motors to give people more time and freedom.
As diehard an urbanist as I am, this blog post is not mainly about urban design. It's November 2026, and I am but one of the masses with thoughts on artificial intelligence. Given the nature of my work as a data scientist at a technology startup, AI has entirely upended my day-to-day experience. I've had to constantly rewire my instincts and think, "would AI do this task more efficiently?" Needing to compete with other businesses, I cannot afford to not use AI. As Claude code emerged this year with superior programming skills to mine, my time spent has rapidly shifted towards waiting for AI to churn and reading its output.
This transformation has been exhausting in unexpected ways. Some mornings, I realize I have been at work for hours and not fully woken up. Many times when faced with a problem requiring thought and research, I have instead redirected the task to Claude. Often its response is perfectly adequate and I'm able to continue my day without activating my brain's deeper neurons. The exertion that writing code or emails used to require has been replaced by passive reading, mostly in reviewing other people's AI-generated code which has increased substantially. Compounding that, I am doing the same with my robot, constantly reading tomes of generic, overly verbose code that come from my account. Reading with an unactivated frontal cortex brings its own form of exhaustion, and eventually I realize I have to snap out of the brain fog. Zooming out from the day-to-day tasks, as our codebase becomes more AI-written, I feel my grip on its structures and linkages slipping away. The promise of AI to give us more time to think has backfired - just as cars led us to commute further, AI has generated more tasks to fill out the workday.
I have enjoyed AI the most for my side hobbies. Recently I was curious to understand OpenAI's proof of a Millenium Prize problem, the Navier-Stokes existence. 15 years removed from graduate school, I am not sharp on reading mathematics papers. But I had some exposure to the Navier-Stokes equations previously via computational fluid dynamics, and by using AI tools to explain the problem and solution to me, I was able to understand the issue well enough within a week to present on it to my colleagues. I spent 15 minutes or so a night breaking down components of the equation and proof, and think about it during the day. Thoughts like "if the proof implies that the equation breaks down, what does that mean physically?" would flutter in, and I'd come back to AI with follow-up questions. On my own, I would have spent much time just figuring out what to read. When I first went to the Wikipedia article for this topic, but it was quite intimidating as I lacked the foundations to make much sense of anything. Using AI reduced so much friction by adjusting to my initial level, and then ramping up as I wanted to know more.
AI accelerated my learning by generating 500 words on Navier-Stokes in a few seconds, but I had to do the hard work by reading and thinking at a steady pace. My brain is limited both in speed (The Unbearable Slowness of Being claims we can process ~ 10 bits/second) and in storage. I can't read that fast and I hit fatigue, and thus I was only able to understand satisfactorily the complicated Millennium Problem over multiple days. AI is able to consolidate the information of the known internet incredibly quickly and effectively, but we are constrained in our ability to absorb this information by biology. Our brains are now the bottleneck, and technology has not meaningfully improved this.
Well what's wrong with that? We can still use AI to supercharge what we feed our brain like I just described. The problem is that this is a choice, and we are primed to choose not to do that. It's effortful to use our brain, and very easy to use AI. As we've seen with many tech products, our default usage is impulsive, dopamine-feeding and short-term oriented. Tech companies saw this in their user data and optimized their products around it. Left to themselves, they will naturally cater to the lowest common denominator, and we will get products that enable us to slop our way through the day. We will get the digital equivalent of cookie-cutter suburbs full of dazed drivers.
Left to the free markets, suburbs have become the natural outcome. It is very easy for private developers to buy land and build houses, while it's nearly impossible for them to create their own public transportation network. Successfully orienting development around mass transit requires top level government planning. It involves well thought-out subway route planning, massive infrastructure buildout, prudent mixed-use zoning, and scale. This allows us to use subway cars to go fast and cover great distances and then finishing the last bit with our feet. Good transient-oriented development allows society to thrive full of face-to-face interactions and productive people.
It is similar for AI in that government can play a major role in imagining a prosperous AI-enabled world. To do so, they will need to apply some reins on the AI companies. I don't know what the right policy solutions are, but one lesson that I've learned from using apps over the last decade is that while usage and addiction may feel like a personal problem, they're heavily shaped by product design. AI products are no different - imagine if there wasn't a button to copy your output from Claude? Even a slightly more friction-inducing interface would reduce slop and encourage us to slow down. Left unchecked, I fear a world where most texts are AI-generated, most thoughts are unoriginal, and no one can make any decision without consulting AI.
However, without removing accountability from the big tech builders, there are personal habits we can install. We can choose to center our brain in our workflow, to use AI less for its output and more as a learning tool. It might seem antiquated or even wasteful in the age of efficiency-maxxing to do tasks "manually", but sometimes it's worth the mental engagement. AI output might always be faster in the short-term, but actual knowledge gain pays off in the long-term. We can choose an Intelligence-Oriented Design, or we can succumb to a life of AI Slopurbia.
In this post, I've consciously resisted letting AI write sentences, even as I am tempted, in this extremely meta moment, to ask it to rephrase this awkward sentence and cut out some commas. Partly I wanted to challenge myself and keep my writing skills sharp, like intentionally walking to the store after driving around all day. The output, this blog post, isn't nearly as valuable as the exercise of fleshing out this analogy and thinking through how I really feel. But I did use AI to bounce ideas, to solidify the Marchetti's Constant analogy, get feedback on structure and maybe generate one image. AI helped a lot, but I had to avoid rushing along. I needed to spend time and let ideas stew overnight before the piece came together. I deliberately tried to utilize intelligence-oriented design as I conceived and wrote about it, lest I am swallowed up in hypocrisy. Hopefully by keeping up this practice, when we are inevitably forced to live in Slopurbia, I can at least choose to live near a bus line.
