3 ms·

While the hardware production capacity is interesting, the main race in mobility is nowadays the AI race. I find it surprisingly hard to get some insight on ho

While the hardware production capacity is interesting, the main race in mobility is nowadays the AI race. I find it surprisingly hard to get some insight on how Chinese companies are doing on that front. The only thing I find is a few videos of western journalists who do a single ride somewhere in China. Does anyone know about sources to get a better insight on how Chinese companies are doing in terms of self-driving software development?

talldayo 2y ago
The hardware production is the race. I don't like BYD, but they've taken full advantage of Tesla's unwillingness to deliver on a truly economic vehicle. Until Tesla takes that crown, the entire AI discussion isn't really even on the table.
- jansan 2y ago
  What I find baffling is that Volkswagen has absolutely nothing to offer in the low price segment. We were lucky to get an eUp! when the price was reasonable, but after they stopped production there is a glaring hole in their portfolio, and of course the Chinese will try to fill that
  - talldayo 2y ago
    Thank goodness all of our hype, investment and subsidies went to one man that doesn't even intend to create human jobs or sell cars to anyone the rich and supremely insecure. Sure was worth the entire free market going gaga for this tasty nothingburger, yesiree.
    - FeepingCreature 2y ago
      I mean, that doesn't really excuse VW's failure.
  - deelowe 2y ago
    I suspect the engineers at traditional automakers HATE electric vehicles and this is likely causing massive delays. The vast vast majority of those engineers and other technical folks have careers which are are utterly dependent on ICE development and support. This sort of transformative shift rarely happens from within. At least not easily.
    - bux93 2y ago
      Probably not the engineers, but the suits. Switching over from ICE is an enormous business problem because it's not just about manufacturing, but the whole after market ecosystem. EVs require less maintenance, but dealerships make most of their money on replacement parts. Getting rid of some uppity ICE engineers (in so far as they're not retiring) is the least of your problems as an ICE automaker.
    - jansan 2y ago
      As an engineer who used to work in the car industry I do not think the engineers hate electric cars. Also, they are not the ones making the decisions. Volkswagen basically bet the farm on electric vehicles. They were late to the party, but went all in, so you cannot blame them for trying to block EVs. I rather suspect that making cheap EVs is just extremely difficult if you intend to make a profit. I bet they lost a good sum of money on every eUp! they produced (here in Germany you could get them for 15,000€ at the right time, which is a steal), so they decided to quit producing it, which IMO is a real shame, because it is such a cool car for inner city commuting.
- Animats 2y ago
  The entire US auto industry is unwilling to deliver on a truly economic vehicle. The "more crap per car" mindset is resulting in huge parking lots full of overpriced unsold cars. The "electric car is a premium product" mindset is obsolete, but continues to show up in US auto pricing. Amusingly, the Ford Transit commercial van is cheaper in electric than in gas-powered. But on the consumer product side, there's still a big price premium for electric, and you usually have to get an excessive 'trim level'. Stellantis screwed up so badly that sales dropped about 40% and the CEO was fired, after collecting the biggest bonus in automotive history for "cutting costs". Betting on "mild hybrids" and extra post-sale fees was a disaster. Nobody in US automotive is investing in solid state batteries the way BYD and CATL are. There's technical risk, but if those things work, IC cars are over. 9 minute charge and a few hundred miles of range.
  - seanmcdirmid 2y ago
    > Nobody in US automotive is investing in solid state batteries the way BYD and CATL are. There's technical risk, but if those things work, IC cars are over. 9 minute charge and a few hundred miles of range. This, even if you embargo the new tech, if the new tech is a lot better, it will create a more broad economic disadvantage in not adopting it.
    - Animats 2y ago
      There's a tipping point coming. Once 9 minute charging works, charging stations work like gas stations, not parking lots. Nobody stays long. It makes sense for legacy gas stations to rip out the fuel pumps and tanks, and put in fast chargers.
      - seanmcdirmid 2y ago
        Gas stations have a lot of infrastructure for catching oil run off, they could put in superchargers, but it would be a horrible use of land. The superchargers can literally go anywhere with power access since EVs don't leak anything other than possibly some coolant. If gas stations become obsolete, they will be subject to rehabilitation before they used for charging.
- ArtTimeInvestor 2y ago
  That might be your personal perspective, but the market's attention is on the AI aspect. Tesla's market cap is higher than that of any other western car company. And not because of their ability to mold metal. In the future, a car without self-driving software will be as useless as a computer without an operating system. Hardware companies will be at the mercy of software companies. All the margins will be made in software. Hardware already has super thin margins nowadays. Nokia didn't fail because of the way they made phone hardware.
  - kavalg 2y ago
    Can you be more specific on the use cases that comprise that? E.g. is it driverless company fleets, enabling elderly to drive, making it possible to e.g. sleep while driving, being fancy etc?
  - ppsreejith 2y ago
    It's possible that self driving can be 'added on' using something like Comma: https://x.com/comma_ai/status/1795596189544722695 https://x.com/comma_ai/status/1795596189544722695
Loic 2y ago
If you do not make a lot of money somewhere (Google ads, Tesla high valuation), it is very hard to have enough money to do the self-driving stuff. So production is important, because if you sell a lot of cars, you have cash flow for R&D.
- ArtTimeInvestor 2y ago
  You expect Tesla to run out of money before they start making money from self driving software?
  - ashoeafoot 2y ago
    yes.
ChumpGPT 2y ago
I'm not sure self-driving is that important in most of the markets BYD will expand into: Mexico, Central and South America, Asia, and where allowed in Europe. I don't think Tesla or any other manufacturers have anything to worry about in the US, Canada, and some EU countries since it will be impossible for BYD to sell cars into those markets profitably.
seb1204 2y ago
We don't need self driving cars for the transport revolution but self driving busses, trains and maybe trucks.
- ArtTimeInvestor 2y ago
  User experience always wins. The best user experience is: You tap a button on your phone Immediately a car arrives You enter it It drives you directly to your destination You exit
  - MrHamburger 2y ago
    Unless car arrives in 3 hours because it is morning peak and everyone wants to get to job.
  - SilverBirch 2y ago
    That won't be the user experience. There is a limited amount of space on the road and functionally unlimited demand. Demand will level off at a point where either "Immediately a car arrives" turns to "After a barely tolerable wait your car arrives" and "It drives you directly to your destination" turns to "It sits in traffic for a barely tolerable amount of time". Either that limits demand or price limits demand in which case the vast majority of people will be priced out in order to keep roads empty enough for your best user experience.
    - ArtTimeInvestor 2y ago
      Let's think it through. If both are available to you, then no matter how crowded the roads will be, a car that drives you directly from A to B will be the better UX than a bus that drives you from somewhere near A to somewhere near B via C, D, E and F. So the car could only be the worse experience if it is hard to get one (Because all the cars near you are occupied by other people) while a bus is easy to get into (Because there is one near you and you still fit in). The vast majority of worldwide rides will continue to be done in individual cars. The extremely crowded situation that would make a bus the better experience is extremely rare. Most rides worldwide are definitely not done in that kind of environment. And the removal of parking spaces and the better routing self-driving cars allow will move even those environments more towards the individual car.

talldayo 2y ago

The hardware production is the race. I don't like BYD, but they've taken full advantage of Tesla's unwillingness to deliver on a truly economic vehicle. Until Tesla takes that crown, the entire AI discussion isn't really even on the table.

jansan 2y ago
What I find baffling is that Volkswagen has absolutely nothing to offer in the low price segment. We were lucky to get an eUp! when the price was reasonable, but after they stopped production there is a glaring hole in their portfolio, and of course the Chinese will try to fill that
- talldayo 2y ago
  Thank goodness all of our hype, investment and subsidies went to one man that doesn't even intend to create human jobs or sell cars to anyone the rich and supremely insecure. Sure was worth the entire free market going gaga for this tasty nothingburger, yesiree.
  - FeepingCreature 2y ago
    I mean, that doesn't really excuse VW's failure.
- deelowe 2y ago
  I suspect the engineers at traditional automakers HATE electric vehicles and this is likely causing massive delays. The vast vast majority of those engineers and other technical folks have careers which are are utterly dependent on ICE development and support. This sort of transformative shift rarely happens from within. At least not easily.
  - bux93 2y ago
    Probably not the engineers, but the suits. Switching over from ICE is an enormous business problem because it's not just about manufacturing, but the whole after market ecosystem. EVs require less maintenance, but dealerships make most of their money on replacement parts. Getting rid of some uppity ICE engineers (in so far as they're not retiring) is the least of your problems as an ICE automaker.
  - jansan 2y ago
    As an engineer who used to work in the car industry I do not think the engineers hate electric cars. Also, they are not the ones making the decisions. Volkswagen basically bet the farm on electric vehicles. They were late to the party, but went all in, so you cannot blame them for trying to block EVs. I rather suspect that making cheap EVs is just extremely difficult if you intend to make a profit. I bet they lost a good sum of money on every eUp! they produced (here in Germany you could get them for 15,000€ at the right time, which is a steal), so they decided to quit producing it, which IMO is a real shame, because it is such a cool car for inner city commuting.
Animats 2y ago
The entire US auto industry is unwilling to deliver on a truly economic vehicle. The "more crap per car" mindset is resulting in huge parking lots full of overpriced unsold cars. The "electric car is a premium product" mindset is obsolete, but continues to show up in US auto pricing. Amusingly, the Ford Transit commercial van is cheaper in electric than in gas-powered. But on the consumer product side, there's still a big price premium for electric, and you usually have to get an excessive 'trim level'. Stellantis screwed up so badly that sales dropped about 40% and the CEO was fired, after collecting the biggest bonus in automotive history for "cutting costs". Betting on "mild hybrids" and extra post-sale fees was a disaster. Nobody in US automotive is investing in solid state batteries the way BYD and CATL are. There's technical risk, but if those things work, IC cars are over. 9 minute charge and a few hundred miles of range.
- seanmcdirmid 2y ago
  > Nobody in US automotive is investing in solid state batteries the way BYD and CATL are. There's technical risk, but if those things work, IC cars are over. 9 minute charge and a few hundred miles of range. This, even if you embargo the new tech, if the new tech is a lot better, it will create a more broad economic disadvantage in not adopting it.
  - Animats 2y ago
    There's a tipping point coming. Once 9 minute charging works, charging stations work like gas stations, not parking lots. Nobody stays long. It makes sense for legacy gas stations to rip out the fuel pumps and tanks, and put in fast chargers.
    - seanmcdirmid 2y ago
      Gas stations have a lot of infrastructure for catching oil run off, they could put in superchargers, but it would be a horrible use of land. The superchargers can literally go anywhere with power access since EVs don't leak anything other than possibly some coolant. If gas stations become obsolete, they will be subject to rehabilitation before they used for charging.
ArtTimeInvestor 2y ago
That might be your personal perspective, but the market's attention is on the AI aspect. Tesla's market cap is higher than that of any other western car company. And not because of their ability to mold metal. In the future, a car without self-driving software will be as useless as a computer without an operating system. Hardware companies will be at the mercy of software companies. All the margins will be made in software. Hardware already has super thin margins nowadays. Nokia didn't fail because of the way they made phone hardware.
- kavalg 2y ago
  Can you be more specific on the use cases that comprise that? E.g. is it driverless company fleets, enabling elderly to drive, making it possible to e.g. sleep while driving, being fancy etc?
- ppsreejith 2y ago
  It's possible that self driving can be 'added on' using something like Comma: https://x.com/comma_ai/status/1795596189544722695 https://x.com/comma_ai/status/1795596189544722695

Loic 2y ago

If you do not make a lot of money somewhere (Google ads, Tesla high valuation), it is very hard to have enough money to do the self-driving stuff. So production is important, because if you sell a lot of cars, you have cash flow for R&D.

ArtTimeInvestor 2y ago
You expect Tesla to run out of money before they start making money from self driving software?
- ashoeafoot 2y ago
  yes.

ChumpGPT 2y ago

I'm not sure self-driving is that important in most of the markets BYD will expand into: Mexico, Central and South America, Asia, and where allowed in Europe. I don't think Tesla or any other manufacturers have anything to worry about in the US, Canada, and some EU countries since it will be impossible for BYD to sell cars into those markets profitably.

seb1204 2y ago

We don't need self driving cars for the transport revolution but self driving busses, trains and maybe trucks.

ArtTimeInvestor 2y ago
User experience always wins. The best user experience is: You tap a button on your phone Immediately a car arrives You enter it It drives you directly to your destination You exit
- MrHamburger 2y ago
  Unless car arrives in 3 hours because it is morning peak and everyone wants to get to job.
- SilverBirch 2y ago
  That won't be the user experience. There is a limited amount of space on the road and functionally unlimited demand. Demand will level off at a point where either "Immediately a car arrives" turns to "After a barely tolerable wait your car arrives" and "It drives you directly to your destination" turns to "It sits in traffic for a barely tolerable amount of time". Either that limits demand or price limits demand in which case the vast majority of people will be priced out in order to keep roads empty enough for your best user experience.
  - ArtTimeInvestor 2y ago
    Let's think it through. If both are available to you, then no matter how crowded the roads will be, a car that drives you directly from A to B will be the better UX than a bus that drives you from somewhere near A to somewhere near B via C, D, E and F. So the car could only be the worse experience if it is hard to get one (Because all the cars near you are occupied by other people) while a bus is easy to get into (Because there is one near you and you still fit in). The vast majority of worldwide rides will continue to be done in individual cars. The extremely crowded situation that would make a bus the better experience is extremely rare. Most rides worldwide are definitely not done in that kind of environment. And the removal of parking spaces and the better routing self-driving cars allow will move even those environments more towards the individual car.