On April 25, 2019, George McGee was behind the wheel of his 2019 Tesla Model S when it slammed into a parked Chevrolet Tahoe on a dark Florida road.

The Tesla was using Autopilot at the time – Tesla’s driver assistance technology that enables a vehicle to steer, accelerate, and brake automatically within its lane.

But this wasn’t just a fender bender.

The force of the collision launched the Tahoe into two pedestrians: Naibel Benavides Leon and Dillon Angulo.

Leon was killed. Angulo was severely injured.

McGee was reportedly speeding and rummaging for his dropped phone in the car when the crash occurred.

Tesla’s system, which the company admits is not fully autonomous and requires an attentive driver, did not disengage.

Last week, a Florida jury found both McGee and Tesla at fault.

McGee bore 67% of the blame. Tesla carried the remaining 33%, for what the jury ruled was a defective product design and failure to warn drivers.

The total damages against Tesla? $243 million.

Given how closely I follow this space, I wanted to share a few thoughts.

This was a terrible accident.

A woman lost her life. A man will carry lifelong injuries.

There’s no sugarcoating that.

But this isn’t the end of self-driving.

If anything, it’s part of the difficult process of progress.

Tesla’s technology is already safer than human drivers by a long shot.

According to Tesla’s Q2 2025 safety report:

• Autopilot-equipped Teslas crash once every 6.69 million miles.

• Non-Autopilot Teslas crash once every 963,000 miles.

That stat isn’t a green light to ignore real concerns.

Autopilot isn’t perfect.

And I’ve long disagreed with Tesla’s decision to rely solely on cameras (known as Tesla Vision) rather than combining cameras with lidar and radar like many competitors do.

Redundancy matters in safety systems.

But a single tragic case, even one this serious, doesn’t mean we throw out the technology altogether.

Aviation had fatal crashes early in its history. So did the first cars. So did elevators.

Progress often comes with painful setbacks.

That’s not justification. It’s reality.

Tesla plans to appeal and they may succeed in reducing the punitive portion of the damages.

Florida law sets a high bar for punitive awards in product liability cases, and Tesla argues that McGee’s own actions (speeding, failing to watch the road, overriding Autopilot with his foot on the accelerator) were solely responsible.

And yet the court made clear: Tesla still carries responsibility.

Autopilot wasn’t designed for the smaller road McGee was on and Tesla doesn’t force disengagement when drivers stop paying attention.

There’s also the name: “Autopilot.”

It evokes the idea of a self-driving car when in fact the driver is still responsible.

That messaging has caused confusion for years.

And it’s not just a branding issue, it’s a legal one.

This verdict may slow things down. It may trigger regulatory changes. It may even push Tesla to rethink parts of its approach.

But what it won’t do is stop the inevitable.

Self-driving is coming.

We’ve seen this same story before.

In 2018, Uber’s autonomous test vehicle hit and killed a pedestrian.

It was a turning point, but it wasn’t the end.

Uber paused its program. Then retooled. Then partnered. Now they’re back on the roads, including here in Austin.

Innovation tends to follow a familiar pattern: early excitement, inevitable setbacks, public backlash, careful refinement, and eventual maturity.

Tesla may be under fire right now, but it’s important to keep the bigger picture in mind.

The technology isn’t flawless.

But when stacked against the reality of human drivers – who are often distracted, drowsy, reckless, or impaired – it’s already showing signs of being safer.

And before long, “better than human” won’t just be a milestone. It’ll be the standard.

So while this moment is undoubtedly serious, when you take a step back, it looks a lot like progress.