There’s no place like safety and the pucker factor

That scrappy Tesla is still in the news for safety issues. NHTSA has requested more data about their Autopilot recall from December. With over 20 crashes since then there are some questions that need to be answered. And the DOJ and SEC are investigating them about their self driving claims. Drunk driving has been increasing but enforcement has declined. We discuss solutions that are available today but that manufacturers and Congress don’t want to act on. Aurora is ready to put their self driving semi’s on the road and this inspires Fred to come up with a new business idea. Michael suggests that the EPA regulate the fire retardants in cars so that we don’t get exposed to horrible chemicals. Plus Recall Roundup.

This Weeks Links

Subscribe using your favorite podcast service:


note: this is a machine generated transcript and may not be completely accurate. This is provided for convience and should not be used for attribution.

Welcome. Good morning, everybody. Ellie meditation. One person speaking at a time, please. Yeah, you ruined my daily meditation. But that’s okay. Let’s You guys ever hear of a company called Tesla? Never did. They’re interesting. My

Michael: Neuralink, my Neuralink implant prevents me from hearing bad things about Tesla.

Anthony: Okay for the next few minutes, Michael won’t be able to hear anything. Investigators with the U. S. National Highway Traffic Safety Administration have concerns about whether the recall remedy that Tesla put out what was this, a few months ago, has worked, because since this Recall has come out.

Tesla has reported 20 crashes since it was sent out in December. Now, to refresh your memory, this is, I guess it was December, we’ve talked about this, where they had to recall every car they ever made, and they’re like, Hey, let’s just put a little text message there that says, Keep your hands on the wheel, or something of that nature.

And a whole bunch of Tesla people are like, It’s not a recall, man! How could it be a recall? It’s just blah blah blah blah blah. Elon’s the greatest. Hopefully he doesn’t fire me. So anyway yeah, Nitsa’s finally wait a second. Maybe your recall, your remedy. Not so good. I did for context and this might be a bit of an ask, but okay.

This recall came out in December. It is currently the beginning of may 20 crashes since then. That seems pretty high.

Fred: Yeah. The nature of the recall, I think it’s important to state was that this determined that. The warning being given to the drivers was inadequate and that the, in particular, the cars were able to be used with no warning outside of the roads for which they were designed, right?

So they could go anywhere and do anything, except in a lot of places where they weren’t supposed to do it. Isn’t that the gist of it, Michael?

Michael: Yeah, just to dial down on that, there are literally five areas where NHTSA was requiring Tesla to put in place some better systems to help drivers, essentially.

One of them was mode confusion. Basically, consumers are a little bit confused, it seems. Or drivers when the Tesla’s in autopilot, when it’s not, they wanted to make sure that the driver knows when they’re when they need to control the vehicle when it’s not in autopilot, they need to pay attention all the time, but their control is somewhat restricted during autopilot operation.

The 2nd, 1 was. Allowing the Tesla’s allow the autopilot to be engaged when they are off highway or not in an area where there’s a highway that has, essentially. Heavy trucks or other vehicles that have stop signs that have to cross the highway at certain times.

We’ve seen how dangerous that is with Tesla. Wants to make wanted to make sure that Tesla’s can only be in it could only enable autopilot while they were on a controlled access highway because of those issues. They also wanted to increase the driver monitoring in the vehicles to make sure that the vehicle knows or the vehicle has some information as to whether or not the driver’s controlling the vehicle after autopilots engaged, if the driver can respond.

And then they had, there were some things like Tesla was going to spend people who got forced out of autopilot because they were basically behaving badly, not the driver monitoring system detected that they weren’t doing things right, or they weren’t holding the steering wheel. And also they wanted to increase the size of the alerts available to the driver.

So almost all of these were. Things that pertain to the driver essentially staying in touch with the road environment and the vehicle instead of, completely vegging out and not paying attention to the driving task.

Anthony: So NHTSA has gone back to Tesla and asked them for data on all. Accidents that have occurred with autopilot engaged and given all of this data that’s happened.

They’re asking for it now, and my question is why didn’t they ask for it months ago, like when this recall happened? Or, why haven’t they been Getting this kind of information the entire time.

Michael: Yeah. I think that the, there’s a couple of answers to that. One is, Tesla’s it’s not easy to get information out of Tesla.

And the big reason behind it is that NHTSA only, the NHTSA doesn’t have a lot of authority to dictate what remedy manufacturers are going to use to repair a safety defect or a noncompliance. That is, if it’s a non compliance, the manufacturer has to bring the vehicle back into compliance.

So there’s a hard line there. With a safety defect, the manufacturer has to do and perform a remedy that, eliminates or significantly reduces the chance of that safety defect occurring. Here, Tesla said, we’re going to do these five things, apparently in October they had some negotiations around this.

These were the five areas that were identified where Tesla said, we’re going to do this and it said, okay. Tesla sent in the recall notice said they were going to do these things. And then, they did some, they didn’t do some, there was one in particular where I think they put in a a function to give a more pronounced slowdown event that alerts drivers when they’re disengaged from autopilot and are fully in control of the vehicle.

But they installed that function in a way that it wasn’t active when it was installed. Drivers literally had to go into their touch screen menu and dial all the way down to turn on that functionality. So essentially, you’re installing a recall fix in the vehicle, but it’s, it doesn’t occur until the consumer goes in and turns it on.

So that’s a pretty big problem. And that’s just one of many here. This is Typical for Tesla, it’s the way they’ve been treating not just this recall, but a number of others. It’s a breadcrumbing strategy. They give NHTSA just enough to let NHTSA think they’re going to do the right thing.

Then they go and do whatever they want to make it look like they’ve done the right thing and There’s always a question whether they did do the right thing. We’ve seen this in a phantom breaking recall and investigation that was going on with Tesla. They put out a supposed remedy for that. And then the phantom breaking complaints actually rose instead of fell.

And this is just the latest example of Tesla, saying it’s going to do something and then not really following up on it. And essentially. Essentially challenging this to call their bluff. And that’s not something that’s has done very well, in my opinion, over the last seven or eight years since we’ve been tracking this issue.

But this latest, questionnaire, it’s an information request that was sent to Tesla, really digs into some of the meat here. It’s asking Tesla for it’s, essentially it’s engineering. Studies showing that these fixes that were proposed and installed are actually going to work.

It’s asking them for the human factors research that was involved in developing these recall remedies. And now we’re going to wait until July 1st and probably a lot longer than that because Tesla’s going to, ask for this entire submission to be redacted, and we’ll probably never see it.

But we’re gonna wait to see what Netsa does. Tesla has until July 1st to respond. And at that point, Netsa, has the authority to say your recall was terrible, and we’re going to fine you a large amount for not complying with Motor Vehicle Safety Act. This one is a wait and see, yet another wait and see situation.


Anthony: from the Washington Post article speaking specifically about NHTSA their risk averse approach to regulation puts drivers at greater risk on the road, Senator Richard Blumenthal, Democrat from Connecticut says, who has long called for more safeguards for driver assistance systems. They live in fear of potential criticism or opposition and that’s just no way for a regulator to view the world.

The Elon Musks of the world are not going to just sit still. NHTSA has not made full use of its powers to protect the public, nor come to Congress asking for new powers to keep up with the rapidly evolving technology, he said. But, can’t Congress just force NHTSA to change their ways? And be like, this is what you guys have to do.

Or is this me being very naive and hoping that the world works in a better way?

Fred: Yeah, it’s telling the grasshopper. It’s so cute. The fact is that most of the people who leave NHTSA as top administrators end up working for the companies that they’ve previously regulated

Anthony: no

Fred: built in incentive for NHTSA top executives To do as little as possible, not that I can impugn any particular person, but, some of the administrative.

I can’t,

Michael: The James Owens, who is the last confirmed. The 2nd to last confirmed administrator and that’s under. The Trump administration went to Nero as one of their directors and, NHTSA under him approved Nero’s application for an exemption from NHTSA standards. So it’s something that’s been going on for many years.

Mark Rosekind went to Zooks. He’s no longer there. And then there’s been, just I would say dozens of NHTSA personnel at high levels who have exited the agency for the industry. It’s something that’s been going on throughout the history of NHTSA, but it’s certainly more pronounced in recent years, the exodus, not to automotive.

Manufacturers or the traditional automotive manufacturers, but to the new tech firms that are, they’re designing a lot of this autonomous stuff.

Anthony: So there’s no rules that can be put in place saying, Hey, you work for a regulator, there’s a, I don’t know, five year period or one year period, even like gardening leave.

You can’t go work for the company you just regulated for a year. Is that anti American? You’re shaking your head at me, Michael. Fred’s looking. It would

Michael: be, it would be great. But, those rules would probably also prevent a significant percentage of former congressmen from leaving Congress and going and doing the same thing, which is something a lot of them do as well.

Anthony: Didn’t they have a lot?

Michael: It’s not getting a lot of traction. Oh, yeah. That’s not getting all attraction. There is There are some rules around lobbying there, but those are also, fairly ineffective and enforced not strictly, or not as strictly as enforced as they should be.

Anthony: Shameful.

Michael: Yeah, plus, getting Congress to pass anything at this point regarding NHTSA that doesn’t come every four years in an infrastructure. Act of some sort is almost impossible right now given the state of our Congress. It’s a deep state.

Anthony: An article in U. S. News hey U. S. News, congrats for still existing.

A Tesla engineer testified in 2022 in a lawsuit over a fatal crash involving autopilot that one of the videos posted in October 2016 intended to show the technology’s potential and did not accurately portray its capabilities at the time. Musk, nevertheless, posted a video on social media writing, Tesla drives itself!

No human input at all! Through urban streets to highway streets, then finds a parking spot. In a conference call with reporters in 2016, Musk described Autopilot as Probably better than a human driver. During an October 2022 call, Musk addressed a forthcoming full self driving upgrade, he said, would allow customers to travel to your work, your friend’s house, to the grocery store, without you touching the wheel.

And this gets to the heart of people buying these cars and thinking, these cars drive themselves. And so what’s coming out here is that U. S. prosecutors and the SEC are saying Hey, are these guys committing fraud on investors? So I guess there’s one thing you can commit fraud against the dumb consumers, but don’t do it against, BlackRock.

Michael: Yeah, it’s this is really sad. We’ve hoped that we’d see action against Tesla by the Department of Justice around simply the fact that Tesla is claiming that these vehicles are X when they’re Y, claiming that they are full self driving or even, Equipped to be full self driving Sunday when we think they’re clearly not.

And, continually beating that drum and, drawing. Drivers into the trap of thinking their vehicles can do more than they can resulting in deaths and injuries. You think, the deaths and injuries are going to spur an investigation in a Tesla for, all of the puffery around their vehicles and that they’re going to get in trouble because of that From NHTSA or from a safety authority, that would be your guess.

But in fact, it’s similar to, how you see, mob, famous mob bosses captured. Now we don’t get them on the, the executions they conducted. We get them on wire fraud or mail fraud or tax fraud. Issues. And that’s what’s happening here. The SEC is going after, or the DOJ in fact is going after, I’m sure the SEC is involved, DOJ is going after Tesla for securities and wire fraud because they’ve misled their investors and consumers about the capabilities of their vehicles.

This is, they’ve been claiming they’re going to be robo taxis forever. I think we discussed last week or the week before about what a joke it is to think that Tesla owners are going to turn en masse their vehicles out on the streets of America and allow other people to ride in them as robo taxis.

It’s silly, but yeah, that’s what Tesla and Elon have been promising the world and specifically their investors with SEC is concerned for many years now. I think they’ve got a great case. And I, I don’t know why it took them so long either. But also, I’m still somewhat disappointed that we’re not seeing, large civil penalties and, perhaps even additional penalties from the DOJ coming to Tesla because the safety issue is involved.

Anthony: Listener, I want to know, and this is not a specific question to Tesla, let’s imagine, would you ever want to buy a car that at night becomes a robo taxi. And if so, what kind of cleaning products do you own and do you use? Do you have a regular cleaning schedule going on? Because that’s really the concern I have.

And while you’re thinking about that and writing in and letting us know, go to autosafety. org and click on donate. Thank you to everyone who has. You’re the greatest.

Michael: What else happens to when you’re, Tesla’s have all those internal cameras. So essentially you’re hopping into a robo taxi where you’re going to be continuously monitored by the vehicle owner.

If they desire,

Anthony: I want this now.

Michael: Yeah, I wonder what, but I wonder what kind of protections Tesla’s putting around, that the folks who are going to be riding in these taxis that will never exist.

Anthony: Oh, they’re going to be riding. All right. Check out my only fans page here. What kind of health insurance do you have, Cesar?

Questions I have in front of me, but let’s go to something more serious. Talking about cars that drive themselves or want to. Let’s talk drunk driving. Because, hey if cars could actually drive themselves, this would be a good thing. But unfortunately, it’s not a thing. From an article in MSN, about 13, 500 people died in alcohol impairment crashes in 2022, according to data released in April by NHTSA.

That is a 33 percent increase above 2019’s toll, and on par with 2021. The last time so many people died as a result of accidents involving intoxicating drivers was 2006. So it seems that drunk driving deaths are on the increase, but enforcement of drunk driving laws are on the decrease. What’s happening

Michael: This is another result of the pandemic that appears to be a negative result.

We, I think that arrests for drunk driving according to the FBI stats were over a million, just over a million in 2019. And then in the first year of the pandemic in 2020 dropped to about 780, 000 and they’ve remained there. Which. To me, it’s pretty clearly demonstrates that there has been less enforcement because we’re seeing just as many, if not more drunk driving deaths and, other situations crashes, we’re seeing more of those.

So there’s pretty clearly been a an increase in drunk driving and drunk driving deaths. Deaths due to the fact that it’s simply not being enforced as it was in previous years. He blamed the pandemic on that. There have been some, other police forces have also been in some circumstances, requested to pull back on their patrolling due to, equity considerations and other reasons.

So there has been a significant drop in enforcement and that, this goes not just for alcohol related enforcement, it goes for speed and traffic enforcement generally where during the pandemic drivers were essentially getting away with murder. We saw the amount of reckless speed and alcohol and drug related crashes skyrocket during the pandemic.

And this is, all part of that same problem.

Anthony: So based on these numbers, this is roughly 25 percent of all fatalities is caused by drunk driving or some sort of impairment, right?

Michael: More than that. Yeah. Yeah, 13, 000 is going to be About a close to a third maybe over a third.

Anthony: Is that right? I don’t think I don’t think you’re I don’t know Fred Is that math correct?

We’re talking 43, 000 roughly deaths a year 13, 000. Is that a third? It’s not a third. I was thinking

Michael: in the 30s, right? Here, let me get my calculator out here.

Anthony: Break out the abacus. The reason I ask that is because this is a problem that you guys have mentioned is solvable today. In that there’s ways to disable the vehicle if the driver is intoxicated.

Correct? This is literally saving, you can save over 13, 000 lives a year. And the technology already exists. It works. Yeah, it,

Michael: We discussed this when we discussed what NITS is doing here, and that’s something we should be mentioning here anyway. Which is they issued an advanced notice of proposed rulemaking on technology that could detect and, Stop vehicles that are being driven by drunks, essentially, or not allow them to ever be started by passively detecting whether or not the driver is drunk.

And, for all, everything looks like that’s going to take a while. A lot longer than we think the technology is, could be put into vehicles a lot faster, but there are manufacturers don’t like it. Manufacturers do not. If you’re the 1st manufacturer selling a car that has technology like that in it, you’re going to lose sales to those people.

They want to drive after drinking. Period. So no manufacturer wants to be the first to put their toe out here. A lot of Americans are against for whatever reason are against that type of technology, monitoring their lives and preventing them from driving while drunk, and they’re not going to buy vehicles that have that technology in them.

So that is a big mountain that we’re all facing in eliminating drunk driving is our own resistance to the technology that could save us.

Fred: I think there’s another factor here, which is that we’re basically in the middle of a gold rush where everybody sees the AVs as the pot of gold at the end of the rainbow.

And so companies are diverting capital into pursuit of this pot of gold rather than into improving the safety of vehicles with technology that’s available today. So there’s no federal mandate that says, Oh, you need to fix the roads in Boston before you set off in your Conestoga for San Francisco to chase the gold.

So that’s, there’s a lot of that. Imped imperative to chase this pot of gold. That’s really delaying the onset and the implementation of technology that’s available today that could be saving a lot of lives on the roads.

Anthony: Manufacturers being I think manufacturers being against that, but let’s look at other things that manufacturers are against saying, Hey, we put this in cars, our sales will go down.

Seatbelts, airbags. Anti lock brakes. These are all I think the

Michael: more important one now is speed, Governors. ISA, or Intelligent Speed Assistance, that could, Prevent speeding. Eliminate speeding. And, there, there aren’t as many fatalities related to speeding as there are related to alcohol every year.

So it’s We have the solutions to eliminate, maybe half Of the fatalities in our roads. And it’s, there’s just so much resistance, not only to my manufacturers getting them into cars, but if you’re a manufacturer, you’re not throwing millions of dollars in research at alcohol prevention or speed prevention, because you don’t want to put it in your cars because it’s going to reduce your sales.

Anthony: I think Budweiser should start making their own car. Budweiser, the car for you, will not drive. It seems the industry is saying, Hey, instead of putting in technology that exists and that’s relatively cheap, let’s get rid of the driver and put in technology that’s really expensive that doesn’t work, called autonomy.

That’s, that seem fair assessment?

Fred: That’s the pot of gold, right?

Michael: It’s

Anthony: a

Michael: pot of gold. That’s actually a bottomless pit that people are throwing money into.

Anthony: No, it would scare me barreling down the highway. Let’s say, I don’t know, an 80, 000 pound drunk driven by a wait, drunk and 80, 000 pound drunk.

Oh my God. That would be Friday. How about an 80, 000 pound truck driven by a drunk? Look, there was something going here. It didn’t work out. It got all tongue tied. Let’s talk about autonomous trucking in Texas. There’s a company, Aurora, we’ve mentioned a few times. And Aurora, for the last decade or so, has been working on Creating autonomous 18 wheelers, essentially, and they’re gonna start putting these out on the road in Texas before the end of the year, and in an article in AP, we have linked here, it says On the test track, reporters saw Aurora’s semis avoid simulations of road obstacles, including pedestrians, a blown tire, even a horse, but the trucks were running at only 35 miles per hour, in a controlled environment, with nothing, nothing unexpected happening.

The trucks are being tested with human safety drivers on Texas freeways at speeds of 65 miles per hour or higher. Autonomous vehicles are coming, and they’re gonna be, hey, let’s put them into 80, 000 pound vehicles, driving at 65 miles per hour. But at least they’re keeping a human on board. So with this they’re not

Michael: though.

Anthony: Right now they are.

Michael: That’s right now they are, but they’re not there when it starts later this year, which is something, I believe there was an, an interview with Chris Armstrong who is the Aurora CEO last year. And he said, next year, we’re going to start, Putting these vehicles on the road in Texas without a driver and that’s what they’re planning to do.

We’re not sure of an exact date, but apparently they’re going to put 20 trucks on the road on I 45 between Dallas and Houston, where I’m sure, they’re, that’s, that’s not like driving through the middle of nowhere. There’s going to be some significant traffic and challenges of dealing with other drivers along that route.

So they’re, that’s, and that’s going to be a challenge to do that and to, bear out over time that these trucks are safe. But it’s coming. And, I think the folks in Texas on I 45 are going to need to be on the lookout, I, thinking about this, I, I pucker up a little when I have to pass a large truck, usually that’s on I 81 where they’re headed south on I 81 near D.

C. where there are, Thousands upon thousands of semi trucks traveling every day. And you get caught in situations where you’re behind one vehicle, that’s passing a truck at, one mile an hour faster than that truck is traveling. And so it might take you 30 seconds to get past a semi truck.

And that’s one of those pucker moments for me where I pay a lot more attention, I think, to my driving environment than I would normally. Would otherwise driving through an interstate and looking over and seeing that there’s no driver in that big truck. I haven’t had that experience yet, but I’m interested to see what that feels like.

I can’t imagine the pucker factor goes down too much.

Fred: Let me submit that if you’re passing a truck on I 81, you are nowhere close to the legal speed limit. Michael, those trucks on that

Anthony: road.

Fred: If you are 60, if you’re driving at the speed limit on that road, you are essentially.

Anthony: I want to know how many phone calls the police will get.

Be like, Oh my God, there’s a truck going down the road. There’s no one driving it. They’re going to get a lot of calls like that. But we have a link to the interview with the CEO of Aurora. And one of the things is, the question they ask is, Can you control the trucks remotely if needed? And part of his answer says, It’ll always operate safely, whether that person is there or not.

But it can ask for help, unless it’s at a cell phone range, which he doesn’t mention. Yeah. And also, he gets optimistic. It’s it’s important to recognize that trucks are involved in something like a half a million accidents here today. So we expect to reduce that number. And I keep expecting to win the lottery.

Hey. But

Michael: there’s some other things to consider too fatal crashes involving heavy trucks, occur in a lot of different areas. And if you look at the stats, only about a quarter of those fatal crashes occur on interstates. So they’re automatically putting themselves under a good spot by doing all interstate highway, and also rural interstates only represent about 12 percent of the fatal crashes involving heavy trucks.

They’re putting themselves, if they’re operating these vehicles only on I 45 between Dallas and Houston, it’d probably be a stretch to call a lot of that aerial rural. Some of it will be, they could be, they’re reducing their chance significantly being involved in fatal crashes, which could help their safety record.

And also, they’re reducing the traffic they’re going to be facing by not going into into directly into cities. So it’s going to be interesting to follow to see, autonomous, are autonomous trucks going to make the transition into, driving into D. C. and pulling up to your local 7 Eleven and making a delivery, or is that going to remain for a long period of time, an area where you only want a human to be in control of the truck?

Fred: I think it’s time to digress into a discussion of Cinematography and how it relates to this situation. Shall we? I’m fascinated. Yes. All right So I’m gonna I’m gonna declare a Tao here We were not ignoring your Tao, oh, yeah Save it for the judge buddy. So let’s have you ever seen a movie called the Wizard of Oz?

Yes And towards the end of that movie, there’s a scene where Dorothy closes her eyes, clicks the heels of her ruby slippers, and says, There’s no place like home. There’s no place like home. And what happens? She ends up going home. Isn’t that nice? I thought the drugs wore off and then she just woke up.

That there may be a factor too. I’m not sure shrooms were big back then, but Hey, you never know. What we’re really talking about with these Aurora and these big trucks is a similar, very similar situation. They’re clicking their heels and saying, there’s no place like safety. There’s no place like safety.

And there, and they’re saying, in Texas, we’re going to be safe. We’re going to be great. Closing their eyes and wishing. But that’s the definition of bullshit. There’s, for example, this AP article you talked about, Anthony, one of the quotes in it is Aurora and other companies are developing the systems.

They argue that after years of testing, they show that the structure will actually be safer than human driven ones. They noted the vehicles, laser and radar sensors can see farther than human eyes The trucks never tire as human drivers do. They never become distracted or impaired by alcohol or drugs. Do you see the rhetorical trick here?

The rhetorical trick here is that they’re investing these trucks with human like qualities and saying, They’re just like humans, only better, right? They can see better than you, they can do everything better than you, they don’t get drunk, they’re just like humans, only better. It’s, but it’s important to realize that none of that’s true and there’s no data to support that can actually happen.

They don’t see in the same way as a human. They, you’ve had children Anthony, or at least one, Michael, and, when the child is about, oh, I’d say 2 years old, you say, look, a fire truck. And they say, yeah, there’s a fire truck or school bus. And they fix it on the school bus. But, computers can’t do that.

Computers look at the world, they look at little bits, then there has to be a very sophisticated computer program that assembles those bits of information into a concept that says school bus or fire truck, and they don’t do that. Now, They’re, so they don’t see in the same way as a human. They never reason and they cannot assemble bits of information into global concepts.

So for example, when you’re driving, you go down the road and you see an accident and then you hear a siren, you think, Oh man, this is bad. Here comes a, here comes an ambulance. I better become alert and look for how to get out of the ambulance’s way. No truck’s going to do that. They never, they don’t really understand things and they can’t associate events with emotions or insights.

When you get on the road and you see a terrible crash, you pucker up, right? You say, oh, shit, this is really bad. Trucks don’t have a pucker factor, as far as I can tell. They don’t think, they do what’s something called statistical inference. Statistical inference is really different than thinking, because statistical inference relies upon what you have known from the past and prior experience and your association that prior experience with a scene that you’re seeing right now.

And it sounds like gobbledygook, because it is. It’s difficult, it’s all fuzzy. There’s no real decision making process in here, except to say it looks something I’ve seen before, so maybe it is, and I’ve got a 25 percent chance of doing it, but I’ve got no better chance of doing anything else, so what the hell, I’ll go ahead and do it.

That’s statistical inference. Most importantly, the trucks never care, and they have no incentive for altruism. When they’re, when they’re going down the road, they never see a vehicle on the other side of the road and say I think I’ll stay on my own side of the road because I don’t want to kill this other person.

That’s beyond the capability of motions or thoughts or protocols that can be invested in a computer. They’re also subject to software and hardware defects. And Anthony, you’ve got no software defects, right? And Michael, you’re, you do perfectly well. The subject of programming errors.

Michael: I discussed my Neuralink defect earlier in the podcast.

Fred: Yeah. All right. That’s not solvable by programming. They also have suffered from initialization errors and the subject to cyber attacks, so yeah, they don’t get drunk, there’s a lot of things that they do that you’d really rather they don’t do. I’m going to refer people to an excellent presentation by Missy Cummings, who is a professor at George Mason University.

That she delivered to the IEEE in Ireland at the end of April, and we’ll put this up on the website, but she goes through these items in detail. And for anybody who’s interested, particularly in the whole idea that computers can see, she has a couple of excellent slides there that show what the computer actually identifies and interprets when it tries to analyze an image of a truck that any infant would recognize as a truck.

But that the computer recognizes as a big person, telephone poles, wheels, and two separate trucks. It’s important that people understand the limitations of what they’re saying. What the industry says is if we put the vehicle on the road that isn’t sufficiently safe, then we aren’t confident in the safety of, then it kills everything else.

Over Ermeson said, and then at another company, Kodiak. So a CEO named Mr. Burnett said at the end of the day, these trucks should be much safer than human drivers. No fooling, they should be safer. Wouldn’t it be nice if you showed us some proof that they are?

Anthony: Hey, why do you keep trying to look behind the curtain?

Stop it. Don’t look behind the curtain.

Fred: That’s the problem with the Wizard of Oz safety approach, isn’t it? It’s, it’s faith based engineering at its best.

Michael: Fred, on that note I wanted to ask something, because we know, Aurora is not Tesla. Aurora is not doing everything wrong, they are trying to do this using a safety cra Safety case framework, which we’ve discussed in the past, and it’s just something we highly recommend they’ve not even on their safety crit case framework page. They’ve nodded to you all 4600. We’ve discussed as a way of, basically bearing out and proving that your vehicles are safe and will remain safe over time.

But ultimately, even with the safety case and the assessment made there, like Fred says, the choice here is to put these vehicles on the road is ultimately going to be made by humans, not the safety case, no matter how robust the safety case is, there’s a choice of when to deploy the vehicles on the public road that’s going to be made.

By, folks who have who are responsible for making the company money, they’re responsible to their investors. And, that’s, an interesting fact here. The Aurora seems to be checking some of the boxes. Many of the boxes, in fact, that, that we’d like a lot of companies to do when they’re introducing this technology into the road.

But, I still don’t feel like we’re supremely confident in this decision by them.

Fred: All right. And we did talk with Aurora and we appreciate that, but as always, the devil is in the details, right? So you have a safety case analysis and that’s a really good thing. But who is evaluating the safety cases, and who’s got the incentive on the safety cases to represent the consumer?

So what are the safety requirements, and how and by whom are they validated? If they have a good case to make at Aurora and Kodiak, why don’t they make that case? Why don’t they, bring in, oh, a red team analysis, like some folks from the Center for Auto Safety who look at what they’ve done, and perhaps, Offer them an independent critique.

How do they define safe? They’re going to, they say the truck’s going to be safer. Safer than what? Safer than their imagination, safer than, no, I don’t know, a herd of random goats flying down the road, driving down the road. What does that really mean? What metrics are they collecting to measure safety and who’s reviewing the data?

Anthony, have you ever seen a train? Yes. It’s called a train, right? If you get hit by a train, you have roughly comparable, Almost. Concerns relative to being hit by a heavy truck. Outcome’s gonna be about the same. Gonna hit by a heavy truck, and you get hit by a train. One of the things about trains that’s interesting is they are confined to their operating design domain, meaning tracks, right?

And you know where the tracks are, you know where the train’s going to go. The train warns you when it’s coming. That has a horn, it has lights, it rumbles it has mechanisms in place to exclude you from getting in front of the train. There are railroad crossings, there are embankments, there are overpasses, underpasses.

There’s a lot of ways the train protects you from the consequences of interaction with the train. With these trucks, there’s nothing like that. They’re going to be going down the road, You’re not even going to know they’re there. You won’t know that they don’t have a human driver. They will not be programmed to avoid anything that they don’t expect.

In the same way that the cruise vehicles in San Francisco were not programmed to avoid traffic cones that were placed on their hood. They completely immobilized the car. There’s a lot of things that are going to happen to these trucks that have not been programmed in. So the idea that you’ve tested them for a year and everything’s okie dokie is good.

It’s completely inadequate. And, there’s been no independent review of these. They’re just chasing the pot of gold. What else can we say about those two questions?

Anthony: You just mentioned that we, as other road users won’t know that they’re in self driving mode. So this is not going to be like Mercedes where Mercedes, they said, Hey, we’re going to have the outside of the car will be teal lights or I think teal lights to let you other drivers know that.

Hey, look, I. No hands. So these trucks are not doing anything like that.

Fred: Not that they’ve talked about. No.

Anthony: Okay. And then the other question, and I think I’ve asked this a few times, this is this is a safety related question, but it’s also what’s going on here with the business case of this. So these things, they just drive on this interstate, right?

So they’re self driving on an interstate. They don’t have.

Fred: I got to stop you. They’re not saying, they are not saying that they’re only going to be used on the interstates. They’re saying they’re going to be used primarily on interstates and certain secondary roads. So they’re leaving that door open to say we’re going to drive them anywhere we care to at any time, day or night.

Michael: And that’s because, you can’t build a shipping facility directly on top of an interstate. They’re going to have to get from the interstate to somewhere else to unload the goods.

Anthony: That was my question. It’s yeah, how are they getting the goods? Is a human driver taking them on the interstate and then pulling off on the shoulder, getting out, hitchhiking home, and then the car is taking over?

Because they who knows? Who the hell knows? It’s not

Michael: Yeah, I would just guessing, I would say they’re going to, drive from the outskirts of Dallas to the outskirts of Houston, and a human driver will be involved in either, in the final delivery of the goods inside of the urban areas.

Fred: There are, but there are companies that are looking to automated vehicles to do that last mile delivery talking about putting a B’s into step fans and things like that. But I want to leave you with 1 analogy. Okay. In my mind, this is a lot like the evolution of the asbestos industry at the beginning of the asbestos industry.

Everybody thought it was an unlawed good, right? This is great stuff. We’re going to spray it in ships. It’s going to keep them from burning. We’re going to have a house covered in tiles and asbestos. That’s right. We’re going to make it into tiles. We’re going to put it on the side of houses. We’re going to use it for insulation in houses.

We’re going to pile it six inches deep in houses. It’ll be an ecological Ecological benefit, we’re going to burn a lot less fuel to heat these houses. It’s going to be wonderful. And, that was a great model for, oh, I don’t know, 50 years or so until they discovered that they are also putting it into talcum powder until they discovered that they were causing mesothelioma and shipyard workers.

And then, oh, by the way, and a lot of other people as well. And, after 50 years, it turned into this huge remediation problem that destroyed human lives, destroyed businesses, destroyed capital. My opinion is we’re looking at the same thing here. We’re looking at. Bullshit driving this capital investment into a gold rush that, near term, we’ll never see the consequences of this long term.

Yeah, it’s going to come up. And by that time, the companies are hoping that we are also invested in the capital associated with these trucks. There are so many of them on the road. We can’t turn back now. You’re telling me, I’m going to have to retire these 100, 000 trucks and go back and hire human beings to drive these trucks.

Remember folks, the reason behind all of this is so they won’t have to pay for drivers. The reason that the companies are doing this is so they will not have to pay for the human beings. Who are driving the trucks, whether safely or not on this on the streets. There is no evidence. There is no hard evidence that these trucks will be any safer than human beings.

And the evidence so far with self driving vehicles is that there are tens of times to some to thousands of times more dangerous.

That’s the experience that we’ve got from Tesla and from crews and the numbers of deaths that are associated with these vehicles while being driven by computers, killing people. So be aware of folks. What they’re doing isn’t criminal yet. But they know it could be, which is why they’re working so hard to identify themselves, particularly in the state legislatures.

They want to push the burden of their mistakes and misconceptions onto their unwitting fellow citizen lab rats. And that’s you and me, folks. That’s the end of TAO for today.

Anthony: Always on a positive note. I, as you were talking, I’m thinking, I don’t see this really taking off. And the reason being is because Michael mentioned that the self driving truck will go to your 7 Eleven With its delivery.

And having spent too many years of my youth working in a 7 Eleven you don’t want that. Because what, say you’re working at your Piggly Wiggly. You get your shipment in there, and when it’s wrong you wanna talk to somebody right away. And who’s that you’re gonna talk to? The driver. Be like, this is missing, this is broken, I didn’t order that, I didn’t order this.

And the driver goes, oh yeah, this broke in the transit and whatnot. Whereas if I’m the store operator saying, hey, this is broken, the company on the other end is there’s no one there to prove that maybe you miscounted, maybe you didn’t see these things.

Fred: And you have to sign the invoice, you have to sign the invoice in the store and then the truck driver’s got to countersign it to say yeah.

Yeah, we missed these things.

Anthony: We counted and whatnot. I don’t, so I don’t see it happening. Cause if not, you’re just having that inventory person driving behind the truck in a van yeah look,

Michael: this is where the AV hover drones come in to automatically fix any problems with your order.

Come on guys. That’s too, by the way.

Anthony: Oh, boy. And I’d have to clarify. Fred, you used the reference in the movie The Wizard of Oz. The correct movie you should have used is 1986’s Maximum Overdrive, where due to a comet, machines became sentient, including trucks, and they chased down, I believe, Emilio Estevez, down the highway with

Fred: I’ll have to put that on my list.

I also want to point out, I also want to point out, by the way, We all pay for hijacking, right? Because it increases the cost of goods. And unless you’re shopping at the stores where they stole the merchandise is being sold, but so how difficult is it? How difficult is it going to be to stop one of these trucks and hijack it?

All you need to do is get in front of it and slow down your car. Great funding model. I’m gonna, I’m gonna edit

Anthony: this out of

Fred: the podcast, this is how we get money. Think of the logic they’ve got, right? The logic is that they’re never gonna run over a car. If you were a thief, all you do is you have one car to block the passing lane, another car that slows down in front of the truck, it’s gonna pull over to the side of the road and stop, then it’s yours.

This is, there, there will be no way, there’ll be no way at all, to control the hijacking of these loads and theft will be an enormous problem. My prediction is great.

Anthony: Okay. So we got to do is we had to get our own 18 wheeler cab and a couple of traffic cones. We block these things off. We figure out how to disconnect the self driving tab and we put on ours and we drive off into the sunset.

Michael: With a bunch of Twinkies.

Fred: There’s standard fifth wheel connectors, right? Pretty easy to detach the load. I, this is just a bad idea whose time has come. That’s amazing. I love it.

Anthony: Let’s hijack a self driving car. No, listeners, don’t do that. We still need your funding. Go to autosafety.

org and click donate. And know where you can get the money to donate? From you hijacking a car. Self driving truck. Oh god. No. Don’t do

Fred: that. Don’t do that.

Anthony: Whatever. He’s winking and nodding. No, he’s not. Hey, let’s talk about how your car may be slowly killing you. We have an article in Fast Company.

This is really a cheerful episode, isn’t it? This is great. An article in Fast Company about the carcinogenics used inside your car for flame retardants. And now this, I can have a personal story. So my sister in law, she created a a mattress company, essentially, and mattresses are soaked in flame retardants, and she didn’t want to do that because, cancer.

And so what she figured out a way to make it so they’d still be flame retardant and not be soaked in carcinogens. And so she told me this story, and I’ve seen footage of it, where she had to go to this lab in, I don’t know, Long Island, and they set the mattress on fire. And they couldn’t set it on fire, but apparently the guys in the lab were like, Oh, thank you.

This is the first one we get to set on fire that we’re not afraid for our lives. Because it’s not covered in these chemicals. Yeah. Why are we still that, cause that new car smell? That’s formaldehyde.

Michael: This is part of the new car smell. Basically, essentially, there are, there’s the Federal Motor Vehicle Safety Standard, FMVSS 302.

And it requires that The fabrics and other materials used in automobiles meet a burn test. Basically, it’s a test that test how fast the material burns. And in order to meet that standard, the NHTSA standard doesn’t require flame retardants. It just requires that you pass the burn test, similar to how a lot of NHTSA standards operate.

They don’t dictate what technology you use to meet the standard. And as it also is in most cases, the manufacturers have chosen the cheapest option here. They can use chemical flame retardants to pass the test versus what they could be doing, which is creating new materials and new designs that are more expensive than flame retardants, but eliminate the risk for cancer or any other negative effects from those chemicals.

There’s been, there’s some a little misinformation around this. I’ve seen some articles say that, this is fault for mandating flame retardants. That’s not what they do. In fact, in my opinion here is that the EPA should be involved if these flame retardant chemicals.

Are linked to cancer or any other type of illness, then the EPA needs to ban these chemicals from use in consumer products. And that hasn’t been done. I don’t think that the solution to this problem lies with Netscape. NHTSA is not equipped to handle issues involving chemical exposure. It’s just not something they’ve done very well with.

You don’t have to go too far back to see examples of that, like the carbon monoxide issues in Ford Explorers where people were being continually exposed to higher levels of carbon monoxide in those vehicles. NHTSA didn’t do a very good job of that. I don’t believe a recall was ever issued in that case.

And it’s been a, a kind of a, it’s been a, It’s You know, although NHTSA’s statutory goal is to make vehicles, ensure that vehicles are as safe as possible, it’s not often that they’re doing so in a way where they’re saying, telling manufacturers what type of chemicals or materials they can use in vehicles.

Typically, they are You know, doing like they are here, setting a standard that manufacturers have to meet and saying you made it however you’d like. A lot of the burden here, I think, needs to be shifted away from NHTSA and towards the EPA, which is allowing these chemicals to continue to be sold and used on cars.

Fred: Another good example of that and a major component of the new car smell is a group of chemicals called patholates. P T H, whatever, ethylates. P H W H,

Michael: right.

Fred: Now, ethylates have been banned from a lot of consumer products, and especially soft plastic that children would tend to put in their mouths because it has some serious neurological impacts and developmental retardation for people.

There is no limit on how much of this can be put into cars. And so when you are enjoying the new car smell for your car, you’re also breathing these folates deep into your lungs. And we’ve mentioned this in the past, I think, but that goes along the same lines. What, there’s a whole soup of chemicals in your car.

That you’d really rather not have around you, some of which smell fairly pleasant that the somebody should get their arms wrapped around. I guess it’s up to the EPA, Michael, but, yeah, there’s a real super chemicals. And we have advocated. that when you get a new car with a new car smell, open your windows and let the outside world enjoy that new car smell so you don’t have to.

Anthony: Are there any manufacturers not using these carcinogens? Because there’s a big movement towards using vegan leathers or bamboo or in sustainable materials. Are they all using this toxic nonsense?

Fred: I believe they are because it’s they, I don’t think they can make the vinyl components like the dashboards and armrests and all those sort of things.

I don’t think it’s possible to make those without using some of the vinyl softeners that phthalates are typically used for. Phthalates are, used in squishy plastic, and you may have noticed that your drink bottles that you’re carrying are a little stiffer now than they were five years ago because they don’t have the same kind of vinyl softeners in them.

Anthony: That’s why they’re stiffer. Goddamn government. Sorry.

Michael: Yeah, there are there are some cars, some child car seat manufacturers who they also required to meet the FMDSS 302 testing. And there are some car seat manufacturers who have. Gotten rid of the flame retarded and their products. It’s some, they’re catering to, a market of concerned parents with those car seats.

That gives them, an advantage over their competitors when they are able to ditch chemicals like that. So I don’t know that we will see a similar action from vehicle manufacturers until either Nitsa. Maybe modifies its burn test to something that flame retardants have a hard time meeting, or the EPA takes an action here, or, maybe NITS can team up like they do on the fuel economy standards and get something done on flame retardants.

Anthony: Or they can move like a lot of mattress manufacturers now and don’t use any of that stuff and they still pass these flame tests. Because I think the flame tests requirements are similar.

Michael: Yeah. And, this, this, if you, I sent a video to Fred and Anthony a couple of weeks ago, that kind of relates here, what these flame retardants are doing is they’re not extinguishing fires.

They are basically slowing down the fires as they enter the passenger compartment. If you have a gas fire outside that starts outside the vehicle around the fuel tank. It’s going to take time for that fire to migrate into the interior of the vehicle. And the longer it takes, the more time people have to escape.

And in the video I sent you guys a couple of weeks ago, it shows someone being rescued from a vehicle on the side of the interstate that’s in an ongoing, very large fire. And the passenger is, or the driver is extricated from the vehicle, just. Seconds before the entire thing is engulfed.

And it was some very brave people who were, essentially dodging some explosions while they were pulling the person from the car. And it shows you just how time sensitive these situations are. And it shows you, why flame retardants are used in vehicles in the first place, because when you see vehicle fires and a vehicle that’s not protected either with fabric or flame retardants, it’s going to go up a lot faster and it’s going to significantly reduce the amount of time that people have to escape. It’s, I think everyone that’s the EPA needs to move carefully here to ensure that while you are eliminating chemical flame retardants from vehicles, that you aren’t also shortening the time that passengers and drivers have to escape from burning vehicles.

Anthony: I’ve got it. I’m replacing my interior with asbestos. That’s right, asbestos. It won’t catch on fire. Now, let’s move into recalls. Starting off this week, Ford Motor Company. Oh, they’re the winner of the recall roundup, aren’t they? 242, 000 plus vehicles. This is the 2022 to 2024 Ford Maverick. They’re having issues with their tail lamps.

Yeah, the operation position of the light is affected, the break and signal lights remain fully functional, but they’re having some diagnostic trouble codes related to inoperative rear, hair, rear, hair, whatever. Wow, wait, T H Y L T A T E S.

Michael: We need to get you a new tongue today. Yeah, essentially there are a lot there, the owners have been reporting that their taillights are out and that’s a significant problem.

It probably brings the vehicles out of compliance with the regulations on tail lamps, but I have no idea at the moment, I think it’s one Oh eight. We also have we also have. Something interesting in this is, the Ford provides some data to the public in this recall announcement that 4, 727 different vehicles have.

diagnostic trouble codes that are related to this issue. And that’s, that was interesting to me because we rarely see that type of data. It’s connected vehicle data that Ford is now collecting from these newer vehicles is producing that essentially they’re beaming this data back to Ford while they’re in the field.

And it’s not owners reporting these problems. It is The vehicle reporting these diagnostic trouble codes to Ford, and then Ford is able to use those diagnostic trouble codes to go out and investigate these issues, probably, and they would be able to get to these issues quicker because they’re not having to wait for warranty claims or wait for consumers to present them with a problem, they could see it in their data.

So I think that’s a good thing. I think it’s good that Ford has is putting this into their vehicles, that they’re able to look at these things quicker than they had in the past. And that’s one of the reasons I wanted to point this recall out besides the fact that it affects virtually every Ford Maverick built the last three years, about 250, 000 vehicles.

Anthony: That’s very cool. That’s a great argument for connected cars. It tells you, hey, this is what’s wrong. I wish computers would do that. Next one. Mercedes Benz 15, 604 vehicles, the 2009 GL 320, the, and 2007 to 2008 320, and they have a whole bunch of little initials after them, and I’m eventually my life is going to end before I get through all of this.

Oh my word. The list goes on and on. This is an issue with their brake boosters. And it comes with a do not drive warning.

Michael: Yeah, there’s a do not drive warning. And this one’s important because these vehicles are not any, anything close to new, the, this is 2006 to 2012 Mercedes. So we’re talking about what can be 18.

19 year old vehicles at this point, some of them. So it’s important to get the information out to owners in this case, because essentially you’re. You could have the connection between the brake pedal and the brake system fail in this. So essentially you’re going to have the, a loss of brakes in the movies where you have no brakes.

And that’s why it’s important for folks not to drive this. These vehicles in the meantime, and it looks like they’re going to be notifying owners in about nine days. So the fix and the the hopefully the fix is coming out soon. I can’t tell exactly. It looks like they’re going to, they have a modification kit for the brake pedal that’s already being produced.

So should be a quick fix for the, all of the Mercedes drivers out there.

Fred: I think one of the former directors of NHTSA has such a car. That probably better check that.

Anthony: All right. Nissan next in line, 9, 645 vehicles. This is the 2024 Nissan Sentras. And they’re having issues with their CVT, the continuously variable transmission.

Now this one’s interesting. It says based on internal Nissan record. Nevermind. I was hoping this was like the Ford phone home, but I realized internal records just means people show up and be like, my car can’t shift. The drive shaft spline could fail to engage the differential gear in the CVT, causing a loss of motive power without prior warning.

If I’m going to lose motive power, I want to know ahead of time.

Michael: Yeah, this is basically they put in a they changed their plant essentially. And when they changed the manufacturing plant, it looks like they did a retrofit. Modification to replace part of the CVT assembly. And ended up with 9, 000 vehicles that have a drive shaft that’s not fully seated, which essentially means you need to get in as soon as you can.

If you have one of these cars and get it fixed. And I think the notification is going to go out in about a little over a month. So owners should start looking for that then.

Anthony: All right. Our last recall. Is Hyundai 13, 511 vehicles. This is 2024 Hyundai Elantra’s Venue’s Kona’s. The exhaust gas recirculation valve assembly may contain a sensor that was contaminated with solder flux.

I think we’ve seen the issue with solder flux quite a bit later. That

Michael: Yeah, I think it was we welding slag. Welding slag is what we look to. Different band. I they’re related. They’re like they’re kissing cousins,

Fred: right?

Michael: But let, Fred, let’s let Fred tell us what the difference is

Fred: Well, solder flux is a chemical that’s used to clean metal before you apply a solder and solder is. Like glue it’s used to attach two metal parts to each other, whereas the welding slag is from melting the native material. Maybe that sounds the same to you guys, but it’s quite a different process.

Anthony: Close enough for us. We have degrees in philosophy. So anyway,

Michael: what happens here is the exhaust gas recirculation valve assembly the sensor gets contaminated with this solder flux, resulting in electrical short, which essentially can make your car stop. And then a loss of power in the middle of the road, obviously a safety issue.

That looks like you’re going to have about a month and a half before owners are able to get that recall performed.

Anthony: And you’ll be notified by first class mail. Oh boy. That’s our show and hopefully your exhaust gas isn’t contaminated. With solder flux one, I recirculated. Yeah.

Don’t recirculate your solder gas, exhaust gas. So whatever I’m done. I’m gone. Anthony, is there no coffee in New York today? I drank coffee. It just wasn’t strong enough. I need a self driving podcast host. Thanks listeners. Five stars. Tell your friends, I apologize for the excessive use of the word Alright, thank you.

Bye, bye, bye.

Fred: Bye.