5 ms·

Robotics for Uber is not a "barely-related technology." For a lot of us, it was always obvious that the core operational model for autonomous cars was Uber-esq

by beambot 11y ago

Robotics for Uber is not a "barely-related technology." For a lot of us, it was always obvious that the core operational model for autonomous cars was Uber-esque. The fact that every major automaker is investing heavily in autonomy should convince you that autonomous cars are an existential threat to Uber. (Not to mention that the driver situation is one of Uber's biggest issues at the moment!)

ClintEhrlich 11y ago
That last point is really fascinating: The employment status of Uber's drivers could have a significant effect on the speed at which autonomous vehicles are introduced in this country. Uber may accumulate a tremendous amount of cash, only to suffer unfavorable legal rulings that jeopardize its current business model. One of the obvious solutions would be to devote that money towards replacing the drivers with automation, so that the entire "independent contractor vs. employee" issue becomes moot.
- lifeisstillgood 11y ago
  The employment status of Ubers drivers is a tiny proportion of the impact autonomous cars will have, and to be honest unfavourable legal rulings is factored into their model (Uber seems to be going after 100% of all taxi regiemes, but if they get locked out of only x % - where x is some unknown factor say around 60% - then they are still a vast global cash cow. But just look at the changes that autonomous cars will have apart from solving Ubers W2 and pension provision: - 1/3 of major city ground space is parking - replace that with new housing, or nice green parks. - we think drones will revolutionise delivery - but put a lock box into every cars trunk and add in clever routing software and see what happens - freight and passenger changed the railways. What happens when my commute is merely part of a longer freight haul. - social effects - an hours commute facing forwards and concentrating suddenly becomes ... Family time? Sleep time? Work time? Sex and drugs and adultery? - ownership - I own my car. I choose the colours. If Uber buys 1 million cars they won't ask me my opinion - or will I rent my car to them when I am not commuting in it? Will I part own with several others I am fairly certain will not be doing sex and drugs on the nice upholstery? - it's so big it's scary. PS - completely agree with the "it's out of the labs into the real world - great !" idea you wrote at top of the page. PPS A more useful answer would be that Uber developing autonomous cars to defend against regulatory difficulties is such a huge leap it cannot be defended as a business strategy and only as a VC style investment. These unicorns are not planning on surviving by growing their current business but by starting up totally new ones. It's as if Barclays Bank said, well, we don't need all you employees, we're moving all our assets over to that 12 line peel script over there
- deegles 11y ago
  Introducing self-driving cars might harm Uber as well. Drivers might leave en-masse in protest since they're being replaced anyways. If Uber can't roll out SDCs fast enough (maybe parts shortages or whatever) they'll be screwed. Now that I think about it, Uber might have to stockpile the SDCs and release them all at once.
  - harryh 11y ago
    That's never happened before when industries have become automated. It's not likely to happen this time.
  - mycelium 11y ago
    Most drivers need the income. They leave en-masse in protest and there's a lot of incentive for poor drivers to defect from the protest and pick up even more fares.
  - tomjen3 11y ago
    If that happens it will only be a short term hit. You can say anything about capitalism you want, but nothing beats it if you need a particular thing and have the money to pay for it.
lifeisstillgood 11y ago
(I'm going to make some flat statements that are not meant to be inflammatory about Uber, but are there to be a bedrock to the next step in the logic chain I think is important - so read the next few lines with that in mind please :-) Uber's core operating model is to forcibly standardise taxi regulations in a huge number of regulatory regimes by appealing over the heads of regulators to the market in the hope that if they move fast enough all the regulators won't gang up on them all at the same time. It's probably going to work in much of the Western world (apart from France :-) Now this is a regulatory play - something that needs a very different core skill set to autonomous cars. Ubers recent purchase of Nokia (?) maps is a clear "related" area because right now today, having better maps than the opposition will yield benefits whether humans or robots drive. There is not much of a chasm to leap However autonomous cars are a leap. The technology is only "slightly" there. The legal, regulatory and business models over autonomous cars are nowhere like settled. Who owns the cars, what power source they will have, who will build them, who will even let Uber use them. These are all undecided questions - no way of monetising them on the scale to justify a billion dollar investment This is not to say I don't see a market here - it's going to be staggeringly huge, as plain as the belly found my middle. But it's not yet a related business - it's not a business yet. So what I mean is that all these unicorns are taking huge amounts of suddenly available private capital, and making R&D bets on vast new industries that have yet to become businesses. It's like building petrol garage forecourts because you see potential in the Model T. It's coming yes, but it's not your traditional way to invest billions. The traditional way is to find a repeatable process like "buy a shop on a corner and sell coffee at five bucks a pop" So, there is a new source of huge capital, feeding a new species of R&D company that have structures and (potential) business models that have not been seen before, thus not taxed, regulated or understood before. So, long winded, but the distance between Uber today and a potential autonomous car taxi company is enourmous, and that distance implies a lot more fundamental change to our society than just an investment decision implies.
seiji 11y ago
Everybody thinks Uber wants to make robotic cars so everybody will just use uber robotic cars. But, why won't people buy their own robotic cars and put them on day time "search" mode for fares that get directly paid back to the owner of the vehicle? Then the vehicle can still return home for evenings/weekends when real people need immediate on-demand driving without waiting to be picked up.
- deelowe 11y ago
  That's lyft's model, no? :-)
- beambot 11y ago
  You could do that... and Uber could be handle all the fulfillment in "search" mode. Heck, there is precedent with "normal" cars already [1]. [1] https://help.getaround.com/hc/en-us/articles/204371664-Guaranteed-earnings-and-30-day-free-trial https://help.getaround.com/hc/en-us/articles/204371664-Guara... Under this model, you're just providing capital investment in a depreciating asset in return for special privileges. You'd have to do a cost-benefit analysis to see how it would work out for you. Personally... I'd leave the capital investment in automobiles to others and just pay surge pricing in the few instances where I need immediate transportation.
- cowsandmilk 11y ago
  why don't people have their home computers automatically join cloud computing clusters during the day so they can be used and paid for computation while the people are at work?
  - robotresearcher 11y ago
    Because the cost of moving the data to the computer is too high. Locality is king. My private car is parked, unused, near potential customers right now.
- untog 11y ago
  The same reason relatively few people AirBnB their homes. There will be an administration overhead of at least some kind - who cleans the car? What happens when someone throws up in it? And so on. I'd say, why will people buy their own robotic cars at all? At least in major cities anyway - you could buy a membership to a program (ala ZipCar, car2go, etc) and just have "a car" on tap. The downsides of ownership (maintenance, etc) will be gone.
Bartweiss 11y ago
I think this gets to the heart of Uber's play. They're up against a seemingly endless string of employment-practices lawsuits, and it increasingly looks like they're not going to win. The way out isn't to become a cheaper taxi company, or suddenly overcome entrenched protectionism in a dozen nations. It's to keep pulling in funding and stalling legal battles for just long enough to render the entire argument irrelevant. Driver background checks? No drivers to check on. Driver insurance? No one to insure, sue the manufacturer if you hit something. Employee benefits for drivers? Fire all the employee-contractors. It's obvious that automation is a huge win for Uber. The question is whether they can live long enough to preserver their first-mover advantage, and whether they can find stop-gaps to cover the interim where autonomous cars still need drivers. I wouldn't be surprised if they let licensed passengers get behind the wheel to comply with laws before full autonomy hits.

by beambot 11y ago

ClintEhrlich 11y ago

That last point is really fascinating: The employment status of Uber's drivers could have a significant effect on the speed at which autonomous vehicles are introduced in this country. Uber may accumulate a tremendous amount of cash, only to suffer unfavorable legal rulings that jeopardize its current business model. One of the obvious solutions would be to devote that money towards replacing the drivers with automation, so that the entire "independent contractor vs. employee" issue becomes moot.

lifeisstillgood 11y ago
The employment status of Ubers drivers is a tiny proportion of the impact autonomous cars will have, and to be honest unfavourable legal rulings is factored into their model (Uber seems to be going after 100% of all taxi regiemes, but if they get locked out of only x % - where x is some unknown factor say around 60% - then they are still a vast global cash cow. But just look at the changes that autonomous cars will have apart from solving Ubers W2 and pension provision: - 1/3 of major city ground space is parking - replace that with new housing, or nice green parks. - we think drones will revolutionise delivery - but put a lock box into every cars trunk and add in clever routing software and see what happens - freight and passenger changed the railways. What happens when my commute is merely part of a longer freight haul. - social effects - an hours commute facing forwards and concentrating suddenly becomes ... Family time? Sleep time? Work time? Sex and drugs and adultery? - ownership - I own my car. I choose the colours. If Uber buys 1 million cars they won't ask me my opinion - or will I rent my car to them when I am not commuting in it? Will I part own with several others I am fairly certain will not be doing sex and drugs on the nice upholstery? - it's so big it's scary. PS - completely agree with the "it's out of the labs into the real world - great !" idea you wrote at top of the page. PPS A more useful answer would be that Uber developing autonomous cars to defend against regulatory difficulties is such a huge leap it cannot be defended as a business strategy and only as a VC style investment. These unicorns are not planning on surviving by growing their current business but by starting up totally new ones. It's as if Barclays Bank said, well, we don't need all you employees, we're moving all our assets over to that 12 line peel script over there
deegles 11y ago
Introducing self-driving cars might harm Uber as well. Drivers might leave en-masse in protest since they're being replaced anyways. If Uber can't roll out SDCs fast enough (maybe parts shortages or whatever) they'll be screwed. Now that I think about it, Uber might have to stockpile the SDCs and release them all at once.
- harryh 11y ago
  That's never happened before when industries have become automated. It's not likely to happen this time.
- mycelium 11y ago
  Most drivers need the income. They leave en-masse in protest and there's a lot of incentive for poor drivers to defect from the protest and pick up even more fares.
- tomjen3 11y ago
  If that happens it will only be a short term hit. You can say anything about capitalism you want, but nothing beats it if you need a particular thing and have the money to pay for it.

lifeisstillgood 11y ago

(I'm going to make some flat statements that are not meant to be inflammatory about Uber, but are there to be a bedrock to the next step in the logic chain I think is important - so read the next few lines with that in mind please :-) Uber's core operating model is to forcibly standardise taxi regulations in a huge number of regulatory regimes by appealing over the heads of regulators to the market in the hope that if they move fast enough all the regulators won't gang up on them all at the same time. It's probably going to work in much of the Western world (apart from France :-) Now this is a regulatory play - something that needs a very different core skill set to autonomous cars. Ubers recent purchase of Nokia (?) maps is a clear "related" area because right now today, having better maps than the opposition will yield benefits whether humans or robots drive. There is not much of a chasm to leap However autonomous cars are a leap. The technology is only "slightly" there. The legal, regulatory and business models over autonomous cars are nowhere like settled. Who owns the cars, what power source they will have, who will build them, who will even let Uber use them. These are all undecided questions - no way of monetising them on the scale to justify a billion dollar investment This is not to say I don't see a market here - it's going to be staggeringly huge, as plain as the belly found my middle. But it's not yet a related business - it's not a business yet. So what I mean is that all these unicorns are taking huge amounts of suddenly available private capital, and making R&D bets on vast new industries that have yet to become businesses. It's like building petrol garage forecourts because you see potential in the Model T. It's coming yes, but it's not your traditional way to invest billions. The traditional way is to find a repeatable process like "buy a shop on a corner and sell coffee at five bucks a pop" So, there is a new source of huge capital, feeding a new species of R&D company that have structures and (potential) business models that have not been seen before, thus not taxed, regulated or understood before. So, long winded, but the distance between Uber today and a potential autonomous car taxi company is enourmous, and that distance implies a lot more fundamental change to our society than just an investment decision implies.

seiji 11y ago

Everybody thinks Uber wants to make robotic cars so everybody will just use uber robotic cars. But, why won't people buy their own robotic cars and put them on day time "search" mode for fares that get directly paid back to the owner of the vehicle? Then the vehicle can still return home for evenings/weekends when real people need immediate on-demand driving without waiting to be picked up.

deelowe 11y ago
That's lyft's model, no? :-)
beambot 11y ago
You could do that... and Uber could be handle all the fulfillment in "search" mode. Heck, there is precedent with "normal" cars already [1]. [1] https://help.getaround.com/hc/en-us/articles/204371664-Guaranteed-earnings-and-30-day-free-trial https://help.getaround.com/hc/en-us/articles/204371664-Guara... Under this model, you're just providing capital investment in a depreciating asset in return for special privileges. You'd have to do a cost-benefit analysis to see how it would work out for you. Personally... I'd leave the capital investment in automobiles to others and just pay surge pricing in the few instances where I need immediate transportation.
cowsandmilk 11y ago
why don't people have their home computers automatically join cloud computing clusters during the day so they can be used and paid for computation while the people are at work?
- robotresearcher 11y ago
  Because the cost of moving the data to the computer is too high. Locality is king. My private car is parked, unused, near potential customers right now.
untog 11y ago
The same reason relatively few people AirBnB their homes. There will be an administration overhead of at least some kind - who cleans the car? What happens when someone throws up in it? And so on. I'd say, why will people buy their own robotic cars at all? At least in major cities anyway - you could buy a membership to a program (ala ZipCar, car2go, etc) and just have "a car" on tap. The downsides of ownership (maintenance, etc) will be gone.

Bartweiss 11y ago

I think this gets to the heart of Uber's play. They're up against a seemingly endless string of employment-practices lawsuits, and it increasingly looks like they're not going to win. The way out isn't to become a cheaper taxi company, or suddenly overcome entrenched protectionism in a dozen nations. It's to keep pulling in funding and stalling legal battles for just long enough to render the entire argument irrelevant. Driver background checks? No drivers to check on. Driver insurance? No one to insure, sue the manufacturer if you hit something. Employee benefits for drivers? Fire all the employee-contractors. It's obvious that automation is a huge win for Uber. The question is whether they can live long enough to preserver their first-mover advantage, and whether they can find stop-gaps to cover the interim where autonomous cars still need drivers. I wouldn't be surprised if they let licensed passengers get behind the wheel to comply with laws before full autonomy hits.