Not to mention substantially less open. I've been using an OpenAI subscription in Pi Agent for a couple weeks now and it's great. And from what I can tell, 5.5 is a heck of a model.
Waymo inflates their prices to be above that of Uber/Lyft because they don't have enough vehicles to meet demand. But their operating costs / mile are lower than that of Uber/Lyft. I'd estimate their internal cost per mile is approx. half that of Uber/Lyft. They pocket the rest because they need to recoup decades of expensive R&D.
There is also no reason to compete with Uber/Lyft on price because they are just leaving money on the table. When Waymo first launched, we saw them try to undercut (Waymo was about 20% cheaper than Uber/Lyft) but now it's about 20% more expensive. People are willing to pay extra for Waymo, so why would they charge less?
The margin on each Waymo ride is currently very, very high. I don't expect Waymo to cut prices until real competition arrives.
It's not clear to me if their costs are lower yet. Waymo's vehicles are rather expensive (estimates for their newer Zeekrs are around $75k each), and they need to pay some number of remote monitors for exceptional situations (as noticed during the recent blackout in San Francisco). They also have to collect tons of data to build & maintain high resolution 3D maps of the areas they operate in. And they have to pay engineers to improve the self-driving software.
Waymo passed 200 million driverless miles in February. If we optimistically assume they're up to 300 million miles now, and every mile was paid for at $10 per mile, that's $3 billion in revenue since they launched. In that same time, Waymo has gotten $27 billion in funding. Of course they haven't spent anywhere close to that amount, and they are optimizing for faster rollout rather than profitability, but the finances aren't as gleaming as one might expect.
I'm sure Waymo will figure out ways to reduce their costs over time, but right now I think they're charging pretty close to what they need to break even.
We're looking at different metrics, you're analyzing the average total cost, while I'm analyzing the marginal cost. Waymo has enormous fixed costs like you mentioned, mapping cities and paying engineers are not cheap, which need to be amortized over a massive self-driving fleet. But those are fixed costs which don't increase with fleet size. Waymo currently operates only ~3000 vehicles, which is not enough to amortize those fixed costs into overall profitability.
What matters most are marginal costs (i.e. how much does it cost for Waymo to add 1 more ride). Looking at marginal costs, Waymo takes in more money than it spends on each ride, so projecting outwards when Waymo operates a large enough fleet, Waymo will be profitable.
Uber/Lyft run enormous fleets of ~2 million vehicles in the US, and that's how they are able to maintain profitability. They can spread their engineering and management expenses over millions of rides.
---
Doing my own math, the marginal costs for Waymo are:
Revenue: Each Waymo vehicle brings in ~$50/hour
Expenses: Waymo must pay for
* Assume the cost of a vehicle is $100k
* Amortized depreciation of the car (assume vehicles need to be fully replaced after ~250,000 miles, vehicles average 25 miles / hour, vehicles need to be fully replaced after 10,000 hours, cost is $10/hour)
* Maintenance (Assume the total cost of maintenance is an additional 25% of the vehicle price, vehicle price is $100k and vehicle lasts 10,000 hours, cost is $2.5/hour). This is likely an underestimate, I didn't model the cost of a mechanic, so this could be as high as $5-7/hour.
* Support (assume 1 support agent can support 10 vehicles, Philippine support agent costs $10/hour, so amortized $1/hour per vehicle)
* Cleaning (needed daily, costs $1/hour per vehicle)
* Datacenter compute for vehicle coordination ($0.50/hour per vehicle)
* Electricity (Assume $2/hour)
10 + 2.5 + 1 + 1 + 0.5 + 2 = $17/hour to operate a Waymo.
In conclusion, the marginal costs for Waymo is very profitable.
Even when just looking at marginal costs, I doubt Waymo is half that of a human driven vehicle. If we assume a robotaxi lasts 200k miles before being retired, then the cost of the vehicle alone ($75k) is 37.5 cents per mile. If a vehicle drives 200 miles a day, that's $5 of electricity (250Wh/mile x 10 cents/kWh), maybe $15 of labor to clean, and the space to park it near downtown ($3/day). That's another 12 cents per mile for a running total of 50 cents per mile. Then factor in maintenance (tires, brakes, suspension, etc) and you're probably close to $1/mile. Then you also need support staff, remote operators (approximately 1 per 50 vehicles, but paid significantly more than Uber drivers), and plenty of compute and storage for the high resolution maps (which must be constantly updated as the environment changes). And none of that includes the R&D costs to improve the vehicles or the self-driving software. Yes many of these costs decrease as fleet size increases, but it'll be a while before it gets below $1/mile. (Nationwide, Uber's rates are $1-2/mile depending on the area.)
There are other considerations as well. For example, available ride shares can scale up/down with demand, while Waymo & competitors will need lots of spare vehicles to satisfy peak demand.
I'm certain autonomous vehicles will eat up the market currently held by Uber/Lyft/Taxis. It's just going to take longer than a lot of people expect.
Those are some good points. Waymo & self-driving cars only makes sense in markets where the cost of paying a human driver is much higher than the amortized cost of operating a self-driving vehicle. As an extreme example, in Bangladesh a driver will work for ~$0.25/hour. Self-driving cars don't make economical sense there, the additional cost of making a vehicle self-driving never pays off vs having a dedicated driver.
Waymo can expand easily in markets like SF and NYC, where drivers are guaranteed a minimum pay rate of $22+ per hour, but will make less and less economic sense in cheaper labor markets.
I believe sooner or later Waymo will start selling their tech package to car companies, who in turn will pass the cost on to customers either as a huge mark up or subscription services. I don't believe Waymo can survive on self-driving taxi alone, the hardware cost is too much to go global.
In San Francisco, it has to be. Because of prop 22, Uber/Lyft must compensate drivers a minimum of $22.40/hr, plus $0.36/mile for vehicle expenses. Waymo doesn't have this cost, so it's effectively ~$25/hr cheaper to operate than Uber/Lyft.
I looked up the numbers - the estimated Uber/Lyft cost per mile in SF is ~$4.50/mi, and Waymo is trending around $1.40/mi (estimated 2025 number).
Engineering costs are capital/fixed costs, they're paid once to develop the technology and don't scale with the number of trips. Operating costs (which is what I'm discussing here) are what it actually costs to run each ride. Waymo's marginal cost per trip doesn't include a chunk of some engineer's salary.
Once there are enough trips, the fixed engineering costs are spread across more and more trips, exponentially trending towards zero, driving the cost per trip even lower.
Where is this estimate? I found a wide range of estimates in my web search, from a per-mile cost of revenue of $2 (meaning a loss of $2 per mile excluding capex), to up to $50/mile.
The Gemini results when I searched for this cited this Reddit post [1] which cites this Reddit post [2], which conveniently gives your $2/mile answer.
Anyway, digging into the Reddit posts which gave your lower-bound number, the reasoning seems very suspect. In particular, the biggest methodological problem is that they use retail price numbers when Waymo is almost certainly getting wholesale prices. So it assumes $110K ($70K for a Jaguar iPace + $40K for sensors and other AV equipment) for the car depreciated over 5 years, but $70K is the retail price for a Jaguar, including dealer markup, distribution, marketing, etc, and when you are buying thousands of them you are almost certainly not paying retail. Likewise, it figured 25c/kwh for electricity, which is retail off-peak PG&E rates, but Google just buys their own solar panels and pays pennies for electricity. The AV equipment figure of $40K was I recall what it cost back in ~2014; the cost of LIDAR has come down dramatically since then and now runs $500-1000/vehicle, so that number should also be suspect. And if vehicle cost is more like $50-60K/year than $110K/year, $7K/year in insurance is way too high. Hell, Google could just self-insure with their $250B in cash, they've got a stronger financial position than every insurer other than Berkshire Hathaway.
It's hard to measure "cheaper" as an end user consumer, the price you pay for the service, because it's very likely they're operating at a loss to gain market share and growth.
Exact same reason why Uber and Lyft were considerably cheaper than taxis in many big cities when they first launched (eg: Lyft in Seattle in 2013/2014), running at a loss, and the pricing has now incrementally grown to become the same as, or even more expensive than traditional meter taxis in some places.
IIRC, in SF they're slightly more expensive before tip, but having ridden in them in SF, LA, and AZ I've always felt they were cheaper. Over the long run, they will probably end up being cheaper from the wholesale perspective since eventually the parts and technology cost will come down with time and scale while human wages will continue to rise.
That said, it doesn't really matter if they're cheaper as long as they're comparable.
The cars are newer and nicer (for now), they're almost always cleaner since they can rotating one car out for cleaning doesn't mean the driver is losing earnings, they're better drivers than the average ride-share driver, you don't feel the need to tip, and I've multiple of my friends who are women call out that they feel safer in them because there's no risk of the driver being creepy (or worse).
I don't think Waymo is trying to win on price right now. I think as long as they just stay somewhat competitive on that front the other benefits will continue to draw in customers.
Alphabet/Google/Waymo is a technology business, with emphasis on business. They're not running a charity. They're in it to make money. If it's a $20 Uber ride to somewhere, they're not going to leave money on the table and charge $10 because they don't have a driver to pay. They're going to charge $22 for the premium experience because they know people will pay that.
Of course, but I never claimed that they would do that. At some point though other autonomous services will enter the market and Waymo will have to compete with them on price. Even if that doesn't happen (which seems incredibly unlikely), they're still competing against public transit and people driving themselves (or privately owned self driving cars). Not having to pay a driver means the floor where they can make a profit is lower.
If we're in a world where human driven Uber's are $30, you're right that Waymo probably won't charge $20 just to be nice, even if it only costs them $10. They might charge $20 though if their data shows that it would 10x the number of riders or if they're also competing with another autonomous taxi company.
That sounds right. Passenger pays for lower risk, etc. The market sees the company making $2 extra and a competitor will see if they can do it for just $1 extra.
Depends on the region, I think. Lyft and Uber partner with them in certain cities, so you transparently are charged the same as a similar ride with a human driver. It's only a better experience than a human driver, though, in my view. No chance of yapping, more privacy, no chance of your driver being a psycho, cars are better maintained.
If a waymo costs $200k (car+sensors+install labor) and drives 200k miles, then amortizing up-front costs alone are about $1/mile. We don't really know what the TCO of a waymo is, and it's possible it could go down with economies of scale. Rideshare drivers can get paid $1-2/mile although it varies a lot.
That's just the current cost. The long term cost structure should be based on cars that come out of the factory with all the right stuff pre-installed. There's a BOM for some extra components; many of which you might already find in some cars. Otherwise it's just another EV. So, long term the extra cost per mile relative to a driver driving the same car should be cents rather than dollars per mile. And of course if there is no driver, some components like manual controls, dashboards, mirrors, etc. actually become redundant as well. So the total BOM might actually be lower long term.
The driver cost is of course the big saving. And they need breaks as well and don't drive 24x7. A robo taxi only has down time when there are no rides, or for charging and maintenance/cleaning.
Mostly what Waymo is doing currently with customized vehicles is not actually super scalable. But it helps at their relatively small current scale. You wouldn't design a custom vehicle + factory for their current growth rate. That becomes more interesting when they start scaling beyond tens of thousands of new vehicles per year. They are probably in the lower thousands currently.
I think they raised close to 20-30B so far. They say they are doing 500K rides per week. At 15$ per ride that adds up to ~390M/year. That's revenue, not profit. But if they could 100x that by rolling out to more and more cities and larger and larger areas, it's going to add up to annual revenues that add up to more than what they raised. That's not going to happen overnight, obviously. But they seem on a path where they are scaling, optimizing, reducing their cost, and growing.
The risks here are mainly that they won't have the market to themselves. Others are doing robo taxi's too and if any of them starts scaling faster and cheaper, Waymo could hit some growth issues. Also, with multiple companies competing, prices per ride would eventually go down. The next five years are going to be interesting.
> A robo taxi only has down time when there are no rides, or for charging and maintenance/cleaning.
It's wierd to see this fantasy of machines on HN, of all places - that they have no downtime, no additional costs - it's only a savings from employing people), and (not said here) they don't make mistakes.
Lots of machines have far more downtime and cost than people. Many have more maintenance hours than operating hours.
Of course they have downtime. They need to charge. Waymo partners with car rental companies for depot operations where they inspect for damage and clean the cars. Damaged cars need to get repaired.
But, a Waymo car has a huge impact compared to gig drivers. Even ones that do it 12 hours a day. The Waymo fleet is maximally available for both rush-hour busy times plus closing time at bars and nightclubs.
I don't know if they've released updated fleet statistics, but in the past I've been shocked at how small it is relative to the visibility and passenger miles. This indicates that downtime isn't a limiting factor.
Gig drivers optimize, or try to, their individual revenue and other preferences the shape their responsiveness to customer demand. In other words it's two sided market among individuals. Waymo isn't. Waymo optimizes across their whole fleet for revenue, presumably, and customer satisfaction.
> a Waymo car has a huge impact compared to gig drivers. Even ones that do it 12 hours a day. The Waymo fleet is maximally available for both rush-hour busy times plus closing time at bars and nightclubs.
What makes you say that there is a difference, the difference is in Waymo's favor, or that it's so large?
Rideshare drivers also know about rush hour and closing time, and rideshare companies adjust availability for those times. There's no reason to think Waymo will handle it's inventory of rides better than Uber.
> Gig drivers optimize, or try to, their individual revenue and other preferences the shape their responsiveness to customer demand. In other words it's two sided market among individuals. Waymo isn't. Waymo optimizes across their whole fleet for revenue, presumably, and customer satisfaction.
That is really a bizarre notion. Why would Waymo's interests be any more aligned with customers than rideshare drivers? Waymo cars are robots operated by a corporation in another state or country. Rideshare drivers are human beings in the same car, who live in the same city or town.
Of course they operate in the same market, and the influences of that market on supply are the same. But the response of gig drivers and autonomous vehicles to these influences and incentives is obviously not the same. It can't be the same.
Gig drivers have range of needs and preferences. Waymo, same algorithm, same car, same user experience all the time. Forums like HN also tends to have a blind spot about consumer preferences of women. These and other reasons are why Waymo has an outsize impact on a market compared to fleet size.
Why would the people who control Waymo be better at these things than the people who control Uber?
Again, the premise seems to be the fantasy that computers are objective, autonomous gods that don't have biases or errors, none of which is true. And again, it's bizarre to encounter that on HN of all places.
It's a very dangerous fantasy that sacrifices individual, rational thought and places all power in the hands of those who control the computers, like a cult.
During peak hours Waymo is more expensive than standard uber/lyft - I don't pay attention to black/premium pricing. Off-peak the price can be comparable. I mainly check because my wife prefers it.
Same here, as someone who doesn't drive much, and is generally a "vulnerable road user". I've seen Waymos drive. When they screw up, it's by stopping dead under an abundance of caution. They never speed. They can spot a ped or cyclist from blocks away. Every time I take an Uber home, the driver is guaranteed to drive 40+ on the 20mph road in front of my house while blasting through crosswalks with people waiting to cross. The data is not really in yet (still not enough miles to really say if they are safer), but they pass the eye test.
Wow, lots of libertarian absolutists up this morning.
Guys, that's all well and good as a philosophy, but you need to integrate your views into the world around you too. When you live in a society that has _decided_ to collectively shoulder health care costs, and assume responsibility for everyone's health, you also may need some ground rules. I know it sucks, because _you_ may have just been born there and you don't really have a choice in what society you live, so that means care needs to be taken, but it doesn't mean there can never be any cost-of-entry.
I guess they should ban all the chippies too. Everyone is unhealthy in their own way and that’s the cost of doing business. Socializing healthcare does not require banning unhealthy behavior. It turns out that money does in fact grow on trees and they can make more because it’s fucking fake and it always has been. How are we going to pay for this!?! You literally create money. Governments do it all the time for missiles .
Cigarettes don't grow out of the ground to be able to be deep fried. Some private enterprise manufactures them for sale.
Just ban the sale of them in the country. They offer no positive for society or humanity whatsoever. Chippies at least have their origins in actual food sustenance.
If some new slow method of societally expensive suicide hit the market, it would get banned quick smart. Cigarettes have only stuck around so long because of legacy and well funded lobbyists and PR / marketing types that have been happy to lie at the cost of millions of lives.
> If some new slow method of societally expensive suicide hit the market, it would get banned quick smart. Cigarettes have only stuck around so long because of legacy and well funded lobbyists and PR / marketing types that have been happy to lie at the cost of millions of lives.
> Nice, let's defend that.
Many discussions about freedom are just marketing and corporate interests in a trench coat.
Guys, it's all well and good as a philosophy, but maybe you should take a second look and reconsider that the state just keeps creeping more and more into your private affairs and is very glad when you believe them when they say things are "for the children" and "for the public good". One day you might find yourself in jail for sharing a meme that is critical of the government in any way.
The ironic part to me is you're making an argument similar to one the libertarian absolutists make - society can't shoulder healthcare costs because then it'll need to start taking responsibility over how healthily people live their lives. Without even taking a position on good or bad of it, if the "you also may need some ground rules" is going to stick, why not also bring in mandatory exercise and ban people from sugar and alcohol too? Be a big win for healthcare costs and do people the power of good.
I actually quite like your comment, it'd be interesting to have the stats on whether the downvoter objected to your tone or if they made the logical inference that this argument undermines universal healthcare and didn't like that.
There isn't really a slope here. If we take your original comment for the justification, then what is your argument for why sugar or alcohol are OK and cigarettes not? Alcohol and cigarettes are basically the same category of goods.
Exercise is maybe a slippery slope because it requires enforcing a positive action, but if we're going to force people to be healthy anyway, why not? In a practical sense, not a theoretical one? If you've got theoretical concerns, why doesn't that apply to cigarettes?
For me the answer is easy: alcohol and sugar in moderation do not have negative effects. They may have few positive ones, and there's the easy argument that 'in moderation' is a rule followed by exactly no one, but cigarettes have no 'safe' level of consumption. Heck, passive smoking can cause lung cancer. You can't passively absorb sugar or alcohol. Sure, alcohol can lead to putting other people in danger, but there are existing laws around that.
Literally nothing in the world would be less fun or good or enjoyable if cigarettes simply no longer existed (unless you're already addicted, and the day that cigarettes disappear will be the first day of the rest of your longer life).
That seems to be a completely different argument. pkulak was saying this was about the cost of healthcare in a society that has decided to handle such costs collectively. If you want to make an argument that this is about the minimum possible harm done by cigarettes that's a bit of a non-sequitur.
Although I will say a minimum possible harm argument is weird on practical grounds. Members of my family have smoked in the past, its done them some theoretical tiny amount of damage that is so close to 0 as to be the same thing. That doesn't require the police to get involved. The harm done by the amount of work to earn the taxes and pay the police was probably greater than the damage done by the smoking.
> Literally nothing in the world would be less fun or good or enjoyable if cigarettes simply no longer existed
That seems ridiculous. Obviously there are people who smoke for pleasure. I know several. You can't just tell them that they aren't having fun and pretend that counts.
Forking and coming back is what I like to do. At this very moment I've got a forked project that I'm actively using and making tiny changes to as things come up in my workflow. In another week or two, when I'm certain that everything is working great and exactly how I want it, I'll file an issue and ask if they are interested in taking in my changes; mostly as a favor to me so I don't have to maintain a fork forever.
I've been running Opus and GLM side-by side for a couple weeks now, and I've been impressed with GLM. I will absolutely agree that it's slow, but if you let it cook, it can be really impressive and absolutely on the level of Opus. Keep in mind, I don't really use AI to build entire services, I'm mostly using it to make small changes or help me find bugs, so the slowness doesn't bother me. Maybe if I set it to make a whole web app and it took 2 days, that would be different.
The big kicker for GLM for me is I can use it in Pi, or whatever harness I like. Even if it was _slightly_ below Opus, and even though it's slower, I prefer it. Maybe Mythos will change everything, but who knows.
Opus is about 7 times more expensive than GLM with API pricing. And since you can only use the Opus subscription plan in CC, you're essentially locked into API pricing for Pi and any other harness.
So you're either paying $1000's for Opus in Pi, or $30/month for GLM in Pi. If the results are mostly equivalent that's an easy choice for most of us.
Perhaps I'm being extremely daft: If the API is 7 times more expensive, then why is it $1000 vs $30? Or is there a GLM subscription one can use with Pi? Certainly not available in my (arguably outdated) Pi.
I'm not the OP, but it's the latter. I'm currently using the "Lite" GLM subscription with OpenCode, for example. I'm not using it very heavily, but I haven't come close to hitting the limits, whereas I burned through my weekly limits with Claude very regularly.
I am using GLM-5.1 in pi.dev through Ollama Cloud. I am able to get by on the $20 plan. I use it a lot and the reset is hourly for sessions and weekly overall. This is the first week I got close to the limit before reset at about 85% used. I am probably using it about 4 hours a day on average 6 or 7 days per week.
Or tell pi to add support for the coding plan directly. That gave me GLM-5.1 support in no time along with support for showing the remaining quota, etc, too.
It also compresses the context at around 100k tokens.
Man, this just bit me too. I started playing with OC over the weekend (in a VM), and the spend was INSANE even though I wasn't doing anything. I don't see this as very useful as an "assistant" that wanders around and anticipates my needs. But I do like the job system, and the ability to make skills, then run them on a schedule or in response to events. But when I looked into what it was doing behind my back, 48 times a day it was packaging up 20K tokens of silly context ("Be a good agent, be helpful, etc, for 30 paragraphs"), shipping it off to the model, and then responding with a single HEARTBEAT_OK.
Luckily you can turn if off pretty easily, but I don't know why it's on by default to begin with. I guess holdover from when people used it with a $20 subscription and didn't care.
reply