Comment by brutusborn
3 years ago
One good reason is to help the AIs make better decisions through greater certainty. For example, if a road sign has machine readable data there is a greater certainty that an AI will interpret it correctly. This could affect safety and ease of implementation.
The question is what value would a city get after, what seems to be, a very expensive process? What value do they get from unlocking semi competent AI cars that's worth the time, effort, and requires dedicated street space and an inconvenience to everyone else who travels a different way? It doesn't seem like it's replacing public transit and if someone needs a private vehicle to go beyond the area that's optimized they'll need a different transportation solution. It seems like a lot of cost for a very small benefit.
A very small benefit for society but potentially a very large benefit for a select few stakeholders.
Musk largely became the richest person on the planet off the back of the promise that nobody else would have decent EV tech (turns out, they do), nobody else would have the battery production capacity (turns out, they do, plus super ironic when Tesla is partnering with Panasonic for battery production), AND the promise of fully self driving cars (which will happen for Tesla immediately after flying cars).
I don't have a good term for this, but it's basically corruption: a few benefit at the cost of everyone else.
> off the back of the promise that nobody else would have decent EV tech (turns out, they do)
That just simply not true. He became one of the richest people in the world when Tesla cracked really high volume productions of EVs at a quite amazing margin. That is what people didn't believe was possible when they did it in 2018.
And at the same time SpaceX, where Musk owns much more stock off, managed to re-usability operational and started to launch Starlink.
> production capacity (turns out, they do, plus super ironic when Tesla is partnering with Panasonic for battery production)
Tesla is by far largest BEV producer in the world so its actually true. And if you think all Tesla does is partnering with Panasonic you have not been paying attention.
Telsa has its own battery production and also is one of the largest costumers of CATL, LG and Panasonic. For quite a while Panasonic and Tesla have co-developed technology that Panasonic can just sell to anybody.
As you can see, Tesla has to be both a producer and a major consumer of most battery companies in the world to achieve the volume they do.
And others can not simply replicate it because the industry is supply constraint.
> I don't have a good term for this, but it's basically corruption
Tesla is the largest BEV producer in the world, and just recently made a larger profit then Ford/GM combined (if I remember correctly). Tesla has industry leading margin and is still the fastest growing car company of any size.
SpaceX is one of the most advanced technology companies in the world and I don't think you will find anybody serious who disagrees with that.
The only argument you have that makes sense is that Musk over-promised Self-Driving. That is certainty true, but I wouldn't call it corruption. And I don't think that today the stock price of Tesla is hugely inflated by this as there is so much pessimism on self-driving.
2 replies →
It's an enormous benefit to society. A huge number of people are killed or seriously injured in car crashes every year. Plus the quality of life benefits and reduction in property damage.
1 reply →
How is my example, street signs, an inconvenience to others or needs dedicated street space? Signs need to be replaced anyway, just add a code at the next replacement.
You're thinking back and white "it's not replacing...". What does that matter? To me what matters is "can we make transport better"
I thought the question was "can we actually make fully autonomous vehicles?" If we have that, the value is pretty obvious.
If a so-called AI needs help reading a road sign then it stands no chance dealing with the rest of it's surroundings.
Then someone spray paints the sign, a taxi parks in the bike lane just behind a box truck blocking half the entrance to where you're going, meanwhile a cyclist rides the other direction directly toward you because human agents are chaotic and cities are full of them. Oh, a racoon!
Less tongue in cheek: I think we should optimize city centers for human scales and activities, rather than sending cars right through them at all. We already sacrifice so much for cars in their current state, if we have to optimize for their mobility even moreso we are going in the wrong direction I feel.
I'm not saying get rid of cars, just keep them out of the denser parts of cities and the CBD. Put them on roads, not streets.
Cars should just be blocked from cities unless very special situations. Removing cars from cities has huge health and economic benefits. Car free city centers are universally better then car invested cities.
Cars where they are still allowed should move significantly slower, on much thinner roads.
Self-Driving will not solve anything for cities. They will solve even less then EVs.
It seems like it would be cheaper and easier to get rid of cars entirely
But not politically palatable :-)
The environment requires too many constraints, in a city it wouldn’t be unusual for someone to vandalize these road signs.
How well would that work during snow, fog, storms, how resilient would it be to vandalism? Sounds extremely fragile.