A glance costs road
I think car touchscreens are quietly making cars worse to drive. Not the whole car. Just the part where you have to use it while moving.
A modern dashboard is a gorgeous pane of glass where the dials and knobs used to be. Dropping the temperature two degrees now means tapping the screen, waiting for the climate menu, finding the slider, and nudging it, eyes down the whole time. A knob in a twenty-year-old car does the same job without your eyes ever leaving the road. One of these is decades newer. It is also worse at the task.
I don’t think the screen is the mistake. I think we designed it for the wrong moment.
Drag the speed. Same panel, three amounts of attention. Nothing here is deleted, it just waits for a safer moment to come back.
The showroom is not the road
A car touchscreen is designed parked. It is drawn parked, demoed on a stand parked, and photographed for the brochure parked. Sitting still, a single flat pane that holds everything looks like generosity. Climate, media, navigation, the massage seats, the ambient lighting, all one tap away on one beautiful surface.
Then you drive it, which is the only thing a car is actually for, and the generosity turns into a tax. Every one of those controls now costs a glance, and a glance costs road.
The real constraint is your attention
Screen size was never the problem. Attention is.
A physical button has an edge your fingertip can find and a position your hand memorises after a day. You reach without looking. Flat glass has none of that. Every control is the same frictionless nothing until you point your eyes at it to aim. The interface that demos beautifully is the one quietly asking you to look down at seventy miles an hour.
We know this cold on a phone. A thumb, a few seconds, someone half-distracted, so you lead with one thing and hide the rest. A moving car is that same constraint with the stakes turned all the way up. Somehow the industry forgot the lesson exactly where it mattered most.
Faster should mean fewer
Here is the rule I would put on the wall. As the driver’s load goes up, the interface should show less, bigger.
That is the prototype above. Parked, it is a full console, small targets, and that is fine, because you can look. Start moving and the secondary things fold away. Get up to speed and you are left with the two controls you actually reach for, temperature and volume, blown up into targets you can hit blind. The rest is not gone. It is waiting for a red light.
With one exception, and the prototype keeps it: the map. Navigation is the reason the screen is up while you are moving, so it stays. But look at the split. You read the map, you do not operate it. The things you do shrink or disappear. The one thing you glance at holds its place. That is the real rule. Hide the controls, keep the guidance.
Most car interfaces do the opposite. They hold every control at the same size and the same place no matter what you are doing, as if idling in a showroom and merging onto a motorway were the same task. They are not even close.
The buttons are coming back
The market is already correcting. Euro NCAP, the body that hands out those five-star safety ratings, now says that from 2026 a car has to put basic controls like hazard lights, wipers, and the horn on physical buttons to score top marks. Volkswagen has admitted its all-touch dashboards went too far and is putting buttons back. Hyundai says it kept the buttons on purpose, for exactly this reason.
That is not a defeat for screens. It is a clue. The controls you use while moving, without looking, want to be physical. The screen is for everything else, the stuff you do parked, where looking is free.
Where this goes
The interesting part is not the button comeback. It is what a car interface becomes once it takes attention seriously.
It gets context. The car already knows your speed, your steering angle, whether it is dark, whether you are mid-turn. All of that should change what the screen offers, the same way the prototype reads one number and thins itself out. Voice stops being the gimmick that mishears you and becomes the honest answer for the things too dangerous to tap. Head-up displays move the glance back to the road instead of down to your lap. The passenger’s half of the screen and the driver’s half stop pretending to be the same surface with the same rules.
And then the constraint flips. The day the car drives itself, the driver becomes a passenger, attention is suddenly free, and the screen you spent a decade simplifying gets to be a living room again. The hard version of this problem and the easy one will live in the same car, switching places, and the interface has to know which one it is in.
The screen isn’t the problem
The screen ate the dashboard because a flat pane of glass is cheaper to build, easier to update, and unbeatable in a showroom. None of that is wrong. What is wrong is treating parked as the only state a car is ever in.
Design for the moving car, the one with a person in it who cannot afford to look, and the screen earns its place back. Faster should mean fewer. Everything else is just a lovely way to make someone look down.