It feels a bit more damning when they share a "talking points" memo to service centers to direct them how to characterize the failures. Add to that, they had internal engineers saying the actual failure rate is much higher than what was being reported, and it seems like it's venturing into fraud territory.

You mean how VW directed their dealerships to deny there was a power problem while also directing their engineering team to come up with a fix?

Sounds pretty similar to me.

Yes. So I’m assuming from your statement, you would agree that Tesla c/should be fine billions of dollars for fraud, just as VW was?

To be clear, I’m not saying Tesla is unique in this regard. I’m saying there are degrees ranging from “good faith engineering mistakes that are handled appropriately” and “fraud”. The Tesla memos and whistleblower claims seem to tilt it closer to the latter.

Agree. But further than fines. They should fix their certification program to include sampling. Not just testing a purpose made testing vehicle.

I agree with this. I wonder if they require their suppliers to be ISO 9001 certified. It's incredibly common in the legacy auto manufacturers. The fact that it wasn't mentioned by Tesla in any of their rebuttals makes me wonder if they deviated from this best practice. If so, they seem to be forcing themselves to relearn the same mistakes of their predecessors.

This certification area is one I've given some thought to. I think, especially in the AV space, there is a need for an improved standardized certification process. But to be fair, the VW diesel scandal followed a standardized EPA test, and they still gamed it, so it has to be carefully thought out.

That is why I think it should be some form of sampling. Where the agency assessing can pick a random vehicle from the factory and test it. It would make gamifying a lot more difficult

Sampling wouldn't have helped in the VW emissions case. All the cars had the same software to detect if it was on the road or at a test stand and adjust the fuel mapping accordingly. Maybe in the future, they'll do road tests. IMO, any time you can test it in a higher fidelity environment, the better. There's a lot that needs to be worked out on the testing side to prevent gaming the system.

To me the issue isn’t that it happened… it’s that it happened over and over and over.

It seems more like a pattern/SOP than a single mistake they’re embarrassed about.

I had a VW Diesel where they lied about the emissions.

I otherwise liked the car so I traded for a newer model gas version. 5 years later the sunroof started leaking. I never, ever, use the sunroof, so it could not be misuse/abuse VW denied any responsibility, and I was told a repair might cost 3,000.

A year later, they admitted no fault and settled on a class action lawsuit for the exact problem.

They now offer repairs under warranty if you get the repair done at a dealership.

The catch: the sunroofs are not obtainable anymore. There's months or a year backlog because they can't make enough spare parts to cover the broken vehicles.

They knew this was a problem, the newer model vehicles have a significantly redesigned sunroof

I've been on the other side of this. I worked for an automotive manufacturer and there was a spectacular no-damage self-acceleration no-brakes-work event.

The manufacturer said they would not comment until there was an investigation.

Internally there was a hilarious amount of gossip.

We the programmers instantly started to wonder who of us made the bug.

Some engineers thought it was impossible. Others lined up the theoretical setup for that to happen.

Turned out the driver faked it ... or so the police report says.

> Toyota had throttle by wire problem they blamed on drivers.

My memory may be failing here, but I thought that the resolution of that was ultimately that it was the drivers at fault.

They recalled 8 million vehicles. There was a number of ways the throttle would fail to disengage, mechanical and electronic fails