Reader feedback on the Tesla vs. Waymo debate; Why my view of Elon Musk has turned negative

In last Thursday's e-mail, I shared the debate I had with my friend Marcelo Lima about the autonomous-driving ventures of Tesla (TSLA) versus Waymo, which is owned by Alphabet (GOOGL).

The debate generated more feedback than anything I've ever written. My readers are such a wonderful source of insight, thoughtful critiques, and encouragement – thank you!

In today's e-mail, I'd like to share the highlights...

Most readers appreciated how Marcelo and I kept the exchange respectful, with some even calling it one of their favorite e-mails of mine. For example, Darrell W. commented:

Thanks for sharing the back-and-forth between you and your friend Marcelo. I am not picking a "winner" in your debate, but want to say that I am extremely impressed that the two of you can debate very strongly – each making good points – and end the debate as friends. You'll likely share a glass of wine or a beer the next time you meet with not a single bridge burned. I really wish we had politicians who could do that. Rather than debating the facts, they very quickly turn to name-calling: racist, communist, transphobe, Stalin, Hitler, sexist, moron, etc. Maybe you guys should teach a class on how to disagree without being a jerk.

Jimmy G. highlighted how autonomous driving will give disabled people their "life back":

A friend of mine has a son who is 21 years old and has suffered throughout his life with a disease that causes him to have violent uncontrolled seizures episodically every day. He has never driven a car in his life until recently when he purchased a Tesla Model Y. When he feels a seizure coming, he simply tells the car to take over. The Tesla has given him his life back.

I think this is one thing that autonomous driving addresses which is never discussed. People with certain disabilities who cannot safely operate a motor vehicle now have that luxury and newfound independence.

As for who "won" the debate, many readers sided with Marcelo – like David B., who gave his reasoning:

First, I certainly enjoy your missives. They are timely, interesting, and provocative.

But I must say that I do not agree with your Tesla analysis and wholeheartedly agree with Marcelo. I own a Tesla Y with full self-driving ("FSD") and it is just remarkable. It takes the stress out of driving. From my home in Columbus, Georgia, I drove through Atlanta traffic (nightmare), D.C. traffic (just awful), and through Baltimore on I-95 toward New York City where I have an apartment on West End Ave – and back, no less.

It was a piece of cake! I cannot express adequately the wonder of FSD. It is magnificent and will extend my ability to drive as I approach the age of 90.

Thanks for your wisdom nonetheless.

Andrew S. did as well – and thinks I'm biased against Tesla CEO Elon Musk:

Very interesting conversation between extremely intelligent and well-read humans. As Marcelo pointed out, I think you have Elon completely wrong – both his motivations and limitless talents. That conversation sure struck me as one between a person with fact, insight, and experience against another with seething hatred of Elon Musk. Every slight compliment of Musk or admission of his talents/success is followed by a biased smear.

I think your arguments regarding valuation and Tesla stock price are on the mark, and I share your concern of a future price drop (by the way, I don't own TSLA). But the stock price doesn't even come close to reflecting the positive impact Musk and his companies have made on the world. When you use your opinion of his arrogance, your political bias, etc. to make your arguments, it does not come across well.

I do understand that this was just a conversation among close friends, but in my opinion, Marcelo clearly won this one.

Vangelis S. agreed:

You let your emotions and feelings towards Musk cloud and bias your judgment against the company...

How would you feel and act if the person running TSLA was someone with your utmost respect, admiration, and sympathy, and he was in complete alignment with your beliefs, values, and political views?

Would you be so biased, in disbelief, and such a naysayer about the company as you are now?

I don't believe my personal feelings about Musk affect my view of Tesla's stock or my analysis of the critical factors that will determine whether it goes up or down – such as the prospects for the company's autonomous driving system and robotaxi business.

To be clear, I continue to think Musk is one of the greatest engineers and entrepreneurs of all time... and that humanity owes him a debt of gratitude.

But my personal opinion of him has changed greatly for the worse because I believe he changed for the worse.

In his youth, Musk was odd yet lovable. But when he had a huge string of wins and became the richest person in the world, things changed...

His big ego morphed into extreme (often malignant) narcissism. His lies and puffery grew dramatically (like in July when he said fully autonomous robotaxis would be serving half of Americans by the end of the year – total absurdity).

He began amplifying and platforming toxic, hateful ideas and people. He expressed glee as he suddenly shuttered USAID, which, as I'm seeing here in Kenya, has led to widespread suffering. More and more allegations of drug use and sexual harassment are coming out. And he lashes out more than he used to at any person or organization (or continent!) that dares to criticize him.

For the sake of humanity, I hope Musk succeeds with his grand ambitions for SpaceX and autonomous driving. But on a personal basis, he has morphed into someone I find hard to like or respect.

Returning to the debate with Marcelo...

Not surprisingly, many readers thought I made stronger arguments – like John H., who wrote:

[Marcelo] isn't taking into account the fact that he and all of those Tesla drivers who are using the self-driving mode are not sitting in the back seat or, if they aren't crazy, sleeping in the driver's seat – although I have seen someone doing that on a freeway in the Bay Area and got my car as far away from that car as possible.

Musk might get his no-driver system approved in some places because he can implicitly threaten or bribe elected officials with his billions. And some regulators might actually think that having a safer system than human drivers is good enough, even if people are still being killed as a result.

Also, what happens if one or more of Tesla's self-driving cameras becomes disabled without it being evident? There's no backup like there is with Waymo.

Rob H. added his thoughts on Waymo versus Tesla:

Like you, I've experienced Waymo (in Phoenix) and was blown away by how quickly the experience shifted from one of anxious novelty to feeling full confidence in the safety... a shift which took less than several minutes as you see how natural the Waymo drive was.

On the debate, I'm with you. While the team at Tesla and Elon are certainly phenomenally impressive geniuses, I don't understand the resistance to incorporate additional safety technology (lidar [light detection and ranging]) when it is readily available. Even if Elon is proven technically correct that lidar is not needed over cameras... what a HUGE liability risk to take on in the eyes of public perception.

Daniel G. made a number of important and informed points:

While I respect Marcelo's conviction, I found his dismissal of the engineering nuances – and his attitude toward the counter-arguments – to be an oversimplification of a highly complex problem. I wanted to share a few technical realities that challenge the "Tesla wins because of more miles" narrative:

1. The "Quality" Argument (Data Density vs. Data Volume): Marcelo's core thesis relies on Tesla's massive fleet mileage. However, in machine learning, not all data is created equal. A mile driven on a Kansas highway offers significantly less training value than a mile driven in downtown San Francisco or Manhattan.

• Tesla: Accumulates billions of miles, but mostly in "low-entropy" environments (highways, straightforward suburbs).
• Waymo: Operates almost exclusively in high-density, complex metropolitan environments (SF, Phoenix, LA).
• Result: Waymo's data is "denser" in edge cases (pedestrians, cyclists, erratic urban traffic). As the density of valuable data points increases, the raw mileage advantage of Tesla shrinks in relevance.

2. The "Ground Truth" Problem (Lidar vs. Vision): Tesla's decision to remove lidar and radar to rely solely on "vision" is a high-risk bet on software over hardware.

• Lidar provides ground truth: It physically measures the distance to an object with laser precision.
• Cameras must guess: Vision systems must infer depth from 2D images. Marcelo argues that "more data" solves this, but training a model on lower-fidelity data (cameras only) is inherently harder than training with precise ground-truth data (lidar and cameras).
• Result: It is a stretch to claim that a vision-only system can match the reliability of a sensor-fusion system (lidar, radar, and cameras) purely through volume.

3. The Eroding Cost Advantage: The primary argument against lidar has historically been cost. However, we are seeing a classic "Wright's Law" curve here. High-performance lidar costs have plummeted from $75,000 to under $1,000 per unit in recent years and continue to drop. As these hardware costs converge toward zero, Tesla's main economic moat (cheaper hardware) erodes. Meanwhile, Waymo's "sensor rich" approach allows them to validate safety faster.

4. Headline Risk and the "March of Nines": Finally, we must consider the societal and political reality. Engineering decisions are driven by logic. Regulatory decisions are driven by public perception.

• The Headline Risk: The autonomous-vehicle industry faces an asymmetric risk profile. A single catastrophic failure involving a fully driverless Tesla could set the entire industry back by years and hammer the stock.
• The Trust Barrier: Interest groups and regulators are looking for reasons to slow adoption. Waymo has spent a decade slowly building a statistical safety case (the "March of Nines" toward 99.9999% reliability). Tesla's "move fast" approach is ill-suited for a regulatory environment that demands perfection, not just "better than average."

We will eventually get to a fully autonomous future, but it will likely be a longer, bumpier, and more regulated road than the "pure tech" optimists like Marcelo expect.

Meanwhile, Kenneth B. disputed the "March of Nines" argument:

Seat belts in cars were mandated on the basis of a roughly 50% reduction in crash fatalities. Airbags were adopted, though the data shows even less improvement in safety.

If there is good evidence that FSD is at least 2x safer than human drivers, there is precedent for not only allowing it, but mandating it – even if Waymo is better (which is arguable).

If the safety improvement is closer to 10x, as some data analyses suggest, rapid regulatory adoption is inevitable. Unlike seat belts, FSD continues to get better, so safety will continue to improve and, for a while at least, exponentially so.

However, Kirk J. noted:

You are spot on in regards to the emotional response to self-driving cars damaging or killing living things... like... ummm... cats? For crying out loud, Waymo got heat over a cat. [See this New York Times article: How Kit Kat Was Killed: Video Shows What a Robot Taxi Couldn't See.]

It is irrational, of course, given that people driving cars kill cats and plenty of other people.

Tesla should use every conceivable sensing device, including lidar, to avoid everything that can and will happen. This is a case where Musk's ego is most likely getting in the way.

Pat D. notes that plaintiffs' lawyers will have a field day with Tesla:

One thing the tech bros don't seem to understand is that the minute fully autonomous Teslas start to cause fatalities, my friends in the plaintiffs' bar will start using words like "trillionaire" and "killer robots" in front of friendly juries. A lot of fortunes are going to be made with this technology... by the trial attorneys.

And Chuck D. agreed:

If a person is killed by a Tesla, the lawyer for the deceased's family will sue because Tesla doesn't have lidar while everyone else does.

I would love to have Musk on the stand: "Mr. Musk, you say lidar is superfluous because cameras work just as well. Then why does the rest of the industry disagree?"

This is followed by: "Mr. Musk, are you asserting that in the event of complete darkness or extreme fog, the cameras work just as good as lidar? Let's run a trial on that."

Mason B. noted:

There's no way to know how real Tesla's FSD safety record is because Tesla doesn't report how often a human interacts to stop a collision.

This begs the question: Why hasn't Tesla released all of its safety-related data, as Waymo has done? If Tesla's internal data showed that FSD was capable of safe, true autonomy, Musk would surely be shouting it from the rooftops.

This post on social platform X highlights the contrast between the two companies' safety reporting:

Electrek Editor-in-Chief Fred Lambert writes that Tesla's latest FSD, while amazing, is far from the true autonomous driving that Waymo has achieved:

My take on FSD has always been that if it were developed by Tesla in a vacuum, it would mostly be celebrated as a great driver-assistance system.

However, the fact that Tesla sold this to customers as something that "has all the hardware to support full autonomy" and will increase in price as the software will improve, until you can eventually go to sleep in the car and wake up at your destination, has changed everything.

Now, not only was Tesla wrong about the hardware and the price, but we also have to compare it to what was promised: unsupervised self-driving.

As long as it is not that, it is a failure. There's no way around it.

I'll let reader Fabian H. have the last (funny) word:

I think the only way to settle this is at dawn on a field with two witnesses. Choose your weapon: nerf gun or light saber.

Best regards,

Whitney

P.S. I welcome your feedback – send me an e-mail by clicking here.

P.P.S. My family and I are spending three days at the beautiful Olepangi Farm, a five-hour drive north of Nairobi, Kenya. We're relaxing, hiking, learning to make ice cream, milking cows, collecting eggs, and feeding and petting rabbits: