Elon Musk $41B offer for Twitter

[durbster]

skwdenyer said:

Just for completeness, Toyota had several throttle recalls. Only 1 related to floor mats; in other cases throttles simply stuck down on their own. IIRC 100 deaths were attributed to that in the US alone. It was a genuinely big deal, but few are decrying Toyotas as intrinsically flawed because of it.

It seems you do not recall correctly because that's not true. It's one of those myths that keeps persisting. Those fatal incidents were found to be mostl likely people hitting the wrong pedal and panicking.

As countless tests have proven, a stuck throttle can be overcome by the brakes so even if there were a mechanical problem that stuck the throttle down, the cars could have stopped.

I also know this for a fact because funnily enough I've experienced it myself (and in a Toyota...coincidentally?). I was driving my Dad's Avensis and as I slowed for some traffic lights the throttle stuck open. It was absolutely no problem to stop the car and then hold it on the brakes while I figured out what was going on.

It turned out the throttle lever just gunked up and got stuck, so I cleaned it up and put a bit of oil on it and it was fine after that.

I told my Dad and he said, "yeah, it does that sometimes".

[h0b0]

The accelerator pedal cover on my Porsche slipped off and cause the pedal to jam. In this case it stopped me from pressing the pedal so a little safer. I have replaced it twice.

[Byker28i]

Found this interesting, cyber truck has a car wash more and that you can reboot it but it takes 5 hours

https://futurism.com/cybertruck-brick-car-wash

He also received a call from Tesla to check on him. The advisor said that "it is a known issue in the Cybertruck that when you do a screen reset, instead of resetting in the standard two minutes, it takes five hours
https://www.theregister.com/2024/04/20/cybertruck_...

Edited by Byker28i on Monday 22 April 18:05

[off_again]

EddieSteadyGo said:

Also, the term "recall" needs to be updated. It no longer means what it used to mean with so many OTA updates and it often gets misrepresented to create a false impression.

No, I disagree. The term 'recall' is valid because it is wrapped up in a whole bunch of legal stuff. From liability to enforcing manufacturers to track what is happening to consumers being heavily notified if there is an open recall on the car. I bought a BMW diesel last year which had an open recall - a part of the emissions control system needed to be replaced. Nothing major, except that it could catch fire.... well, only in extreme situations. But the dealer couldnt get the recall done themselves. Liability has to be passed to the owner and hence it needs to get done.

Even simple things like making sure you get the recall letter if there is change of owner! It is wrapped up in a whole load of stuff. There have been some pretty horrific recalls, such as the airbags not going off, or going off badly on some Hondas etc. And there are some simple ones that can be fixed OTA. Thats OK, you do what you need to for the fix. But the legal framework needs to stay.

Some of Tesla's recent recalls have been for frankly stupid stuff. Some have been voluntary too. Thats OK too. But the requirement to get them fixed is on the manufacturer (and them alone because they have no dealers). There is no false impression though - they MUST get done, consumers have a degree of protection AND you cant hide the recall by passing ownership. I like this and it seems to work fairly well.

Maybe we can have better clarity on what the recall covers and how it is handled, but to change the name or phrase would be wrong. That would allow manufacturers to manipulate the system and ultimately screw consumers - and unfortunately we have a track record of manufacturers doing the wrong thing in the past.

[off_again]

rscott said:

Very odd. If the problem with the loose pedal was caused by an unapproved lubricant, why are they fixing it with a rivet?

https://www.teslarati.com/tesla-cybertruck-acceler...

That does seem to be the case. A rivet and its done.

I did hear about a recall for Rivian recently too. Seems that certain models may have had a loose suspension joint. Didnt affect that many vehicles, but I did hear about the recall itself. Service tech turns up in a van, goes under the truck and checks the torque on a few bolts. Confirms its all OK and its done. A minute at best?

It might be small and it might be insignificant, but sometimes things go wrong with complex systems like vehicles. Having a process to track vehicles affected and to confirm / rectify the fault is good. We might not like the result or how elegant it might be, but here in the US, there arent that many consumer protection laws, and this is one that I think we should keep, if not get tighter.

[hidetheelephants]

h0b0 said:

The accelerator pedal cover on my Porsche slipped off and cause the pedal to jam. In this case it stopped me from pressing the pedal so a little safer. I have replaced it twice.

Why are manufacturers ill-fitting these carlos fandango pedal covers? What's wrong with a bit of moulded rubber?

[Klippie]

hidetheelephants said:

carlos fandango

Your showing your age there now boy...Hahaha..!!!

[Byker28i]

Always a good thing for Tesla to ask shareholders to pay Musk $56m as they lay off staff, especially as a Judge said no
https://abcnews.go.com/US/wireStory/tesla-sharehol...

[h0b0]

Byker28i said:

Always a good thing for Tesla to ask shareholders to pay Musk $56B as they lay off staff, especially as a Judge said no
https://abcnews.go.com/US/wireStory/tesla-sharehol...

[off_again]

h0b0 said:

Byker28i said:

Always a good thing for Tesla to ask shareholders to pay Musk $56B as they lay off staff, especially as a Judge said no
https://abcnews.go.com/US/wireStory/tesla-sharehol...

Yeah, the optics look really bad on that. On one hand, lets lay off 10-15% of worldwide staff and cut costs dramatically. Cut prices of models and features and attempt to stem the bleed that they seem to be under. While on the other hand, demand that he needs to be paid $56bn! Seriously? The biggest pay package ever in corporate history and one that Musk is doing a very bad job in justifying. Ouch, maybe he could have waited 6 months after the dust has settled?

Of course, there always has to be an ulterior motive to these things. I read an article the other day and it basically covered why Musk NEEDS $56bn. With the interest payments for Twitter due and the revenue continuing to drop (from what I understand), he's on the hook for this. He's massively over leveraged at the moment and I am sure it wont be long before his co-investors in Twitter will be wanting to see some money back. If the Tesla stock keeps dropping and his overall wealth drops too, he's really stuck and there is a risk that he can go from the top 5 richest in the world to bottom 100 and no real way to get out of the hole.

There are a lot of 'ifs' with this and its not clear on some of the details, but what is public does make a lot of sense. And couple that with a frequently reported strategy of Musk of risking it all, it does fit. Biggest climb and fall? There is a distinct possibility that this might happen.

[robscot]

off_again said:

he's really stuck and there is a risk that he can go from the top 5 richest in the world to bottom 100 and no real way to get out of the hole.

Spot on, either scenario would not shock me.

[off_again]

Tesla in the news again - this was going to happen at some point (and being a motorcyclist, I knew it was a matter of time):

https://jalopnik.com/tesla-driver-who-trusted-auto...

Idiot driver puts trust in a system that Elon claims is perfect, hits and kills a motorcyclist at speed. Utter, utter moron. I really hope that the Wash State police throw the book at him and he spends time in jail for that. Interestingly he was given a $100,000 bail, which he posted. Thats pretty harsh, though he did kill someone.

But a little digging in on the whole FSA thing - not sure if anyone caught some of the latest news on the new release? I cant remember the version, 12.6 or something? Anyway, its being hyped up by the fan boys with lots and lots of comments on Twitter saying that its perfect and much better. Except its not.

Videos of the latest version not navigating roundabouts, clipping curbs on reasonable corners and trashing wheels (notoriously thin and soft). Then you have the MULTIPLE videos of it driving straight through stop signs without even attempting to brake! And the last one I saw yesterday, a large trailer being pulled by a truck - trailer was full of garden waste and clearly was full. It seemed to be legal, but the Tesla simply could not see it. Nothing. Of course, with the absence of LIDAR or anything similar, the reliance on video only was easily fooled....

And on similar but non Tesla self driving cars - anyone seen the video of the guy with the T-shirt with a stop sign on it who keeps getting Waymo cars to stop in the middle of the street? Its funny, but it also illustrates that we still have a LONG way to go. The good news is that the Waymo Jag's do stop, but this is going to get very annoying when you are trying to get home after a night out.

[EddieSteadyGo]

off_again said:

...
But a little digging in on the whole FSA thing - not sure if anyone caught some of the latest news on the new release? I cant remember the version, 12.6 or something? Anyway, its being hyped up by the fan boys with lots and lots of comments on Twitter saying that its perfect and much better. Except its not.

Videos of the latest version not navigating roundabouts, clipping curbs on reasonable corners and trashing wheels (notoriously thin and soft). Then you have the MULTIPLE videos of it driving straight through stop signs without even attempting to brake! And the last one I saw yesterday, a large trailer being pulled by a truck - trailer was full of garden waste and clearly was full. It seemed to be legal, but the Tesla simply could not see it. Nothing. Of course, with the absence of LIDAR or anything similar, the reliance on video only was easily fooled....
...

I'd suggest being very wary of some of the FSD clips. There are quite a few people who have made long, unedited video for a long periods of time, which show FSD warts and all. I'd say they are mostly trust-worthy. Even so, some will only upload the "good" trips where they had zero interventions. However, the consensus on version 12 I've seen (the "end to end" AI version of the software) is that is a major improvement.

I've also seen some of those clips you mention, and they were clearly faked, probably to create clicks/views.

[soupdragon1]

EddieSteadyGo said:

off_again said:

...
But a little digging in on the whole FSA thing - not sure if anyone caught some of the latest news on the new release? I cant remember the version, 12.6 or something? Anyway, its being hyped up by the fan boys with lots and lots of comments on Twitter saying that its perfect and much better. Except its not.

Videos of the latest version not navigating roundabouts, clipping curbs on reasonable corners and trashing wheels (notoriously thin and soft). Then you have the MULTIPLE videos of it driving straight through stop signs without even attempting to brake! And the last one I saw yesterday, a large trailer being pulled by a truck - trailer was full of garden waste and clearly was full. It seemed to be legal, but the Tesla simply could not see it. Nothing. Of course, with the absence of LIDAR or anything similar, the reliance on video only was easily fooled....
...

I'd suggest being very wary of some of the FSD clips. There are quite a few people who have made long, unedited video for a long periods of time, which show FSD warts and all. I'd say they are mostly trust-worthy. Even so, some will only upload the "good" trips where they had zero interventions. However, the consensus on version 12 I've seen (the "end to end" AI version of the software) is that is a major improvement.

I've also seen some of those clips you mention, and they were clearly faked, probably to create clicks/views.

The good thing about FSD right now is that it's still unpredictable in certain scenarios. So drivers are on alert, rightly so, just in case.

I'm not sure how Tesla will navigate this problem if they get to the point where FSD is generally excellent. FSD will be quite dangerous at that point, simply due to over confidence of the 'supervisor'.

I wonder what the strategy is if they get there. The better it gets, the more dangerous it becomes sounds weird, but that's the reality they might face if left in the hands of regular Joes. Regular Joes who paid up to $15k for the privilege too. How can Tesla remove something that they paid for? Or do Tesla just let them keep it, knowing that some of them could be in a fatal accident as a result?

It's a dilemma.

[off_again]

soupdragon1 said:

Regular Joes who paid up to $15k for the privilege too. How can Tesla remove something that they paid for?

$15k? No, its $12k. Nope, its $8k. Or maybe you want to pay monthly.... it was $199 a month, now its $99...

Personally I would be pissed if I paid $15k for something that you can now get for $8k.

[EddieSteadyGo]

off_again said:

$15k? No, its $12k. Nope, its $8k. Or maybe you want to pay monthly.... it was $199 a month, now its $99...

Personally I would be pissed if I paid $15k for something that you can now get for $8k.

That's one area where Tesla seems happy to burn bridges with its customers. I'd be treated all those early adopters who over-paid for FSD (and have so far had little benefit) much, much better.

However, if we could get the current FSD v12.3.5 here in the UK, I think it is easily worth $99/month.

[soupdragon1]

EddieSteadyGo said:

off_again said:

$15k? No, its $12k. Nope, its $8k. Or maybe you want to pay monthly.... it was $199 a month, now its $99...

Personally I would be pissed if I paid $15k for something that you can now get for $8k.

That's one area where Tesla seems happy to burn bridges with its customers. I'd be treated all those early adopters who over-paid for FSD (and have so far had little benefit) much, much better.

However, if we could get the current FSD v12.3.5 here in the UK, I think it is easily worth $99/month.

I would actually buy a Tesla and pay $99 a month for FSD too, for proper hands free driving.

I don't think its worth it today though. To be constantly on high alert isn't a great driving experience IMV, and I would prefer to do it myself. So many cars have L2 autonomy these days, and 1 pedal driving, no pedal driving, steering assist, auto traffic progression, FSD in its current iteration is worse than those basic systems in terms of overall driving experience.

Tesla are ahead on capability, but with that capability comes unpredictability. With the lesser systems available from other automakers, you get predictability, which makes the overall package better, compared to FSD. IMO of course. Others will disagree, and I get that. Capability plus predictability is the key measure of success, rather than just straight capability.

[EddieSteadyGo]

soupdragon1 said:

I would actually buy a Tesla and pay $99 a month for FSD too, for proper hands free driving.

I don't think its worth it today though. To be constantly on high alert isn't a great driving experience IMV, and I would prefer to do it myself. So many cars have L2 autonomy these days, and 1 pedal driving, no pedal driving, steering assist, auto traffic progression, FSD in its current iteration is worse than those basic systems in terms of overall driving experience.

Tesla are ahead on capability, but with that capability comes unpredictability. With the lesser systems available from other automakers, you get predictability, which makes the overall package better, compared to FSD. IMO of course. Others will disagree, and I get that. Capability plus predictability is the key measure of success, rather than just straight capability.

If you were only say driving in towns/cities and only ever using the current V12 version, I would probably agree, as you do still need to be hyper-vigilant. However, add in some of our motorways, and dual carriageways, where it is much more reliable, and I think overall it would make the more 'average' journey much easier.

However, it is the new architecture, and the new way of training their neural net which I think makes it interesting. Going back a couple of years, I didn't think they would solve FSD. The rate of progress was slow, and they made little improvement getting to the edge-cases. Now they have removed most of the 'rules' based codebase and replaced it with a neural net, and they are training it on 10 second video clips, the rate of learning has accelerated.

Musk was saying last night he already has 35,000 H100s processing video on training FSD (and I think they cost circa $30k each just for the cards). And by the end of year, he says they will have 85,000 cards. That is gigantic computing power. And they are collecting over 10 million miles of training data every day.

So I think we are going to see what happened with chess computers. Based in the early 1990's most people thought it was impossible for a computer to ever beat a top Grandmaster. Then Deep Blue did it and that opened the floodgates. Now something like Alphazero is invincible against and human player.

They can continue to collect edge-case video data automatically. And it continues to incrementally improve. At some point it is going to pass the average human driver.

I should probably add, just for fun, I decided before the earning calls to buy ATM call options on 300 TSLA shares with a 450 day expiry date. Just keeping in mind our prior discussion about the share price this time next year, I thought I should probably put some money where my mouth is! But that does mean I'm technically 'long' so I am now somewhat biased (although hopefully not too much).

[tangerine_sedge]

Linky : https://www.theregister.com/2024/04/24/tesla_q1_20...

Sales fell 9%, net profit down 55%, and production numbers fell, hence why we saw the 10% cut in workforce.

If it was any other business the shares would drop and Elon would be looking for another job, but he throws out a few promises about a new cheaper model and humanoid robots and the share price goes up. At some point these never-never promises will have to be delivered...

[soupdragon1]

EddieSteadyGo said:

If you were only say driving in towns/cities and only ever using the current V12 version, I would probably agree, as you do still need to be hyper-vigilant. However, add in some of our motorways, and dual carriageways, where it is much more reliable, and I think overall it would make the more 'average' journey much easier.

I was doing that autonomously 8 years ago with Volvo Pilot assist. Leaving work, hit the button at the start of the motorway, the car drove me to my home town, I took over at the exit. Loads of cars can do that now. So I wouldn't be paying $99 a month when you can get it for free from other automakers.

EddieSteadyGo said:

However, it is the new architecture, and the new way of training their neural net which I think makes it interesting. Going back a couple of years, I didn't think they would solve FSD. The rate of progress was slow, and they made little improvement getting to the edge-cases. Now they have removed most of the 'rules' based codebase and replaced it with a neural net, and they are training it on 10 second video clips, the rate of learning has accelerated.

Musk was saying last night he already has 35,000 H100s processing video on training FSD (and I think they cost circa $30k each just for the cards). And by the end of year, he says they will have 85,000 cards. That is gigantic computing power. And they are collecting over 10 million miles of training data every day.

So I think we are going to see what happened with chess computers. Based in the early 1990's most people thought it was impossible for a computer to ever beat a top Grandmaster. Then Deep Blue did it and that opened the floodgates. Now something like Alphazero is invincible against and human player.

They can continue to collect edge-case video data automatically. And it continues to incrementally improve. At some point it is going to pass the average human driver.

I should probably add, just for fun, I decided before the earning calls to buy ATM call options on 300 TSLA shares with a 450 day expiry date. Just keeping in mind our prior discussion about the share price this time next year, I thought I should probably put some money where my mouth is! But that does mean I'm technically 'long' so I am now somewhat biased (although hopefully not too much).

I agree that there is enormous room for improvement via technology, AI etc. The only way is upwards as far as that's concerned.

Elon is talking out of both sides of his mouth though. Designing a new robotaxi? Why? He said last night that there are millions of robotaxi ready cars already in the Tesla customer fleet, and they can jump in and out of their ride hailing app, like a homeowner can jump in and out of Air BNB rentals. So why build a new generation robotaxi when, to paraphrase Elon, FSD software is the differentiator between Tesla and other car makers.

If FSD is the golden goose, why build the Goose a brand new nest, when its already got millions to choose from? If the new platform is going to be dedicated to the Robotaxi, but its too much of a capital risk for the M2, why risk that capital in the 1st place, just for Robotaxi, when all the existing Teslas are robotaxi ready anyway?

When you try and join the strategy dots, they don't make any sense. All I can smell is BS to be perfectly honest.

Reuters leaked the news of the shelving of the M2, share price dips hard, then immediately, we get the Robotaxi 8/8 announcement. Then investors still aren't happy about the vacuum of information, why no M2, so Elon fills the gaps in last night and says there is an M2, but not as we know it.

And that's before we even talk about hardware, and vision only. The minute the weather is unfavourable, there is no robotaxi, even with perfect software. Its a fundamental flaw, vision only. I guess that's why the 8/8 robotaxi makes sense - better hardware.....but....then he says there will be millions of robotaxis from the existing fleet on the roads today. So maybe Robotaxi isn't getting new hardware?? Its nonsensical BS. The dots don't join up in a logical manner. Its feels like seat of the pants whimsical BS.