Every weekend, my good buddy Hiten Shah and I head to The Coffee Movement’s Balboa Street location. They make a nice pour-over. And while it’s about a half-hour from my apartment, the coffee is worth the drive, the time, and the money. It’s a long enough drive for us to talk about life, work, and, most importantly, new ideas about technology. Sometimes, if our schedules sync up, we even go twice a week. What doesn’t change is where we meet. It’s the same location. We drive the same route. The only variable is the time of day we go for our coffee.

On the way to the café recently, Hiten switched the car into self-driving mode. Tesla does a remarkably good job pilotingitself, especially on crowded streets. It’s far more careful and attentive than some of my fellow human drivers. As we drove, I wondered aloud: Why can’t Tesla learn our favored route? Why can’t it pick up our weekly driving patterns and store that information in the system? It doesn’t have to upload anything to the cloud; it could just keep it local. Wheneverit detects the start and destination, it should automatically follow our preferred route, adjusting only for traffic.

Surely, there’s a way to add a personalization layer that makes maps more effective. And why stop with Tesla? So whycan’t Apple Maps or Google Maps do it? This shouldn’t be too difficult. Unless there’s some regulatory reason they’re not permitted to do so.

Personalization isn’t a feature; it is a way of acknowledging that life has texture. My own thinking around this is no different from Neil Postman’s, who, in Technopoly, pointed out that “technological change is not additive; it is ecological.” I’ve written before that data without empathy and understanding is meaningless. The efficiency and convenience of technology are what deaden our human experience. Without personalization, we’re reduced to mere data entries.

Personalized mapping looked like the future when Google bought Waze; yet Google Maps remains only somewhat personalized. It is rarely enough to be genuinely useful. And Apple, as a company, gets a failing grade even on the most basic personalization, so we shouldn’t expect much from Apple Maps.

I’m trying to highlight a core tension between algorithmic efficiency and user personalization. Mapping systems often funnel users onto algorithmically determined routes that prioritize speed and major roadways. I suppose this is a deliberate design choice. These algorithms are optimized for the lowest common denominator: moving the greatestnumber of people to their destinations as quickly as possible, typically via routes the system deems most efficient.

I recently met Rodney Brooks, one of the giants of AI and robotics. He shared a little story from his ride over to see me:“I came here in an Uber this morning,” he told me, “and I asked the guy, at one point, what street we were on. He had no clue. He said, ‘I just follow it.’” (Read my interview with Rodney Brooks over on CrazyStupidTech.com.)

We’ve already become machine-idiots, following blindly. We keep taking the routes handed to us by mapping systems, and we’ve stopped even trying to find our own way. No wonder the algorithms jerk us around.

I’m often confounded when Uber drivers take freeway detours, even when city streets would be faster. Lacking local street knowledge, they inadvertently reinforce the system’s biases, feeding it more of the same data it then uses to directfuture users. With deeper, more contextual understanding of real-world scenarios and user intent, that wouldn’t happen; we’d move beyond simply adhering to a prescribed, albeit “fastest,” route.

Despite increasingly sophisticated edge computing in our devices and vehicles, true personalization remains elusive, pointing to more fundamental limits. Even today’s artificial intelligence systems are hampered by this absence.

“We shape our tools, and thereafter our tools shape us,” John Culkin once observed. In our post-algorithmic age, theopposite seems true: the tools built to serve us end up reshaping us. In our pursuit of efficiency, we’ve ceded the subtlety of choice; the small human quirks that make each journey ours are sacrificed to the machine idiocy we’ve come to call convenience.

No one put it better than the late Steve Jobs.

“Technology alone is not enough — it’s technology married with liberal arts, married with the humanities, that yields us the results that make our hearts sing.”

Personalization might seem, at first, like just a technology feature. Yet today’s maps don’t recognize the weekly coffee shop trip as ritual, and algorithms don’t sense the rhythm of a drive shared between friends.

It reflects who we are, and it’s an opportunity to make technology and machine intelligence more human, more personal. Until our systems understand that, they’ll keep optimizing our lives into neat, predictable boxes even as the messier, more meaningful routes go unlived.

October 29, 2025. San Francisco

Editor’s Note: It is a sign of the times that a substantial number of you are taking advantage of newsletters to read what I publish. I have made a subtle change to the website: we are now publishing as a “newsletter.” If you are a subscriber, nothing will change. If you are not, please consider subscribing — and tell your friends and colleagues to do the same. In doing so, you will be supporting my writing. My goal is to publish up to three times a week — including a weekly list of my reading recommendations. (RSS feed is unchanged.)