Federal safety investigators have spent more than a year pulling apart what happens when Tesla's Autopilot meets a distracted human being.
What they keep finding is not a technology problem so much as a trust problem.
Drivers are climbing into the passenger seat, filming themselves, reading books, and letting the car do the work it was never designed to do alone.
The National Highway Traffic Safety Administration has documented hundreds of crashes tied to Autopilot and Full Self-Driving features, including dozens of fatalities.
In many of those cases, the driver's hands were off the wheel for extended stretches, eyes off the road, attention somewhere else entirely.
Tesla's own marketing has done the software no favors.
The name "Autopilot" borrows from aviation, where pilots still sit at the controls and monitor everything. "Full Self-Driving" goes further, implying a capability that does not yet exist in any car you can buy.
Owners hear those words and reasonably assume the vehicle has things handled.
Then they test that assumption at 70 miles per hour.
The company has responded with cabin cameras and steering-wheel torque sensors that watch for inattention.
But a nagging chime is not a substitute for genuine accountability.
Drivers have learned to defeat the sensors with weighted objects and well-timed tugs, treating safety interlocks like a puzzle to solve rather than a warning to heed.
This is where the story stops being about one automaker and becomes a mirror.
Americans have spent two decades outsourcing attention to devices.
We scroll at stoplights, answer texts in movie theaters, and half-watch our kids while a phone glows in our palm.
Handing the wheel to software is simply the logical endpoint of a culture that has decided focus is optional.
The regulatory picture is a mess of overlapping agencies, delayed reports, and investigations that stretch on for years while cars keep selling.
Drivers get a notification, tap "install," and go right back to trusting the machine.
There is no moment of reckoning, just a slow accumulation of crash reports that most people never read.
What makes this genuinely troubling is how ordinary the failures are.
They are commuters on familiar highways, in good weather, doing something they had done a hundred times before.
The technology worked often enough to feel reliable, and reliability is exactly what makes people stop paying attention.
Car manufacturers across the board are pushing toward more automation, not less.
Every brand wants the headline, the feature list, the sense of being ahead.
The incentives all point one direction, and none of them point toward telling customers the truth: that this stuff is a beta test running on public roads with your family in the back seat.
Our take: A car that drives itself most of the time is more dangerous than one that never does, because it trains us to stop watching.
The fix is not more cameras or louder chimes.
Final Thoughts
It is honesty about what these systems actually do, and a willingness to treat inattention behind the wheel as the serious choice it has always been.