They point out that these rideshare apps can detect what your phone's battery % is, and the TOS says they may access that info.
To test the allegation, they used two phones with different battery levels and compared the prices (at the same time, & with same destination), but did not find pricing differences.
They reached out to Uber who made a statement:
> Uber does not use personalized pricing nor is battery life a factor in determining fares. Pricing is optimized for the network as a whole, not any one individual rider or driver. [1]
Here is the reference Uber provided which describes how prices are set, and how other marketplace functionality works. [2]
If Uber was tripling prices when the battery level fell below 20% (like this Tweet) then more people would have noticed it by now.
I think this is an obvious case of someone attributing market-rate fluctuations to the only metric they can personally see with their own eyes: Their battery level. They can’t see Uber’s network of available drivers and other factors that influence the rate, so they overweight the only data point they can see.
This is similar to the phenomenon where people have a verbal discussion about a topic and then coincidentally see an ad for that product in an app later. When we see 10s or 100s of ads per day, it’s inevitable they we’ll randomly see an ad for something we’ve been discussing. Yet some people see these coincidences and wrongly conclude that their phones are listening to their every word, using AI to parse it out (without draining their battery), and then sending those keywords to an ad server for targeting. It has been debunked many times over but it’s still a commonly held belief that all of your phone apps are listening to your conversations.
It can be difficult to convince some people that coincidences are really just coincidences. Especially in this era where Big Tech has an incredible evil reputation in the journalism and media we consume.
I was in this exact situation last weekend. 3% battery, lost a couple drivers to cancellation (going Marina to Marin), and thought I was done for. Would’ve paid any price, but what they were quoting me felt incredibly reasonable the entire time, and around bar closing time on a Friday, no less. ($60 to San Anselmo)
My theory is different from yours but only slightly. Companies have no use for recording your voice.
1. Audio streaming is expensive and requires lots of storage, even if ephemeral, to be done widely.
2. Natural Language Parsing and transcription is fraught with errors. Go read your visual voicemail sometime.
This, to me, is enough to close the book on this theory.
What is entirely possible is this:
Two individuals have a discussion. This discussion results is one of the individuals searching for a product in a store or online (some stores, like a famous hardware store) can detect proximity to certain items. If they can correlate interest based on a linked (or brokered) profile then they have ascertained that person A is engaged with said item. Targeted advertising is such that it is possible for you to be shown items based on your own interests, but it's also possible you're part of a larger network. Again, all a hypothesis and one hard to prove at that.
This is similar to the incredibly common meme that Instagram and other social media apps are secretly using the microphone to spy on you for advertising purposes.
I don't think it's that people are stupid. Blindly trusting these apps to not take advantage or mishandle their access to personal data is also naive.
It's possible to do business without scanning your customer's faces, tracing their contacts, selling details of their transactions to aggregators, etc ... where does the line get drawn?
And that doesn't even address data that is "accidentally" collected.
Nah. You think that you have never mentioned or searched those things, but in reality you're just starting to recognize them as soon as your call ends because your brain has been primed to pay more attention.
This is the same phenomenon as driving with other people and suddenly asking out loud "Hey let's find such or such type of car". Whatever the make (within reason), you'll probably start finding out a bunch of them. That's because you start paying attention, of course, not because there's suddenly a surge of people driving them at the exact time you're looking for them.
There are other ways to accomplish that without a microphone. Hint: there was another party on your call other than yourself, and their behavior (on the web for example, where you may be linked as friends on the same social media sites) can provide signals you are unaware of, that suggest things about their friends including you.
This should be easy to prove by reverse engineering the WhatsApp application.
If someone could come up with proof of this, they'd probably be set for life in one way or another. Perhaps they could monetize the information by collaborating with short-sellers, or simply kickstart a career using the massive media attention.
If they did it even with a 0.001% chance of it happening that would still be enough to earn them extra profit. They aren't going to be stupid enough to do it every single time.
No, that's what I meant to link. Melee damage was accidentally doubled for 2 whole weeks and not and nobody noticed. Those players play extremely close attention to stats and such.
The post goes on to talk about how and why many changes and behaviours can go unnoticed, and I think the whole episode disproves the parent posters initial point.
I think you're agreeing with OP, namely that it's not battery usage that is causing the price hike, it's time of day that's causing it and the battery life is correlated.
I've heard so many people say they think iPhones are 'listening to them talk' because they'll talk about some product then go on Instagram/Google and see an advertisement for that type of product. Obviously they were just targeted really well and missed some other data connection.
Humans are good at pattern matching to notice trends but non-technical people always tend to blame something higher up in the stack, like the device in their hand. Or jump to something more malicious by the all-knowing powerful big tech companies.
they don't. I think you are just misunderstanding the comment. The poster was saying people might assume its related to battery level since it is correlated to time of day, which does have an effect on prices.
He was making a comment about wrongly thinking causation from the situation
I mean they were greyballing city employees in some cities -- I don't know if that's 'personal pricing' but it's certainly an adversarial personalization feature
VW during dieselgate had the ability to detect the testbench. I might build this here if I were concerned about getting caught. Uber built a remote kill switch for their tax docs in canada, this isn't unprecedented for them.
also nobody denies location is used to personalize pricing, so wtf are they talking about. customizing the user's experience based on location is 100% personalization?
the problem here is less the accusation than the total lack of price transparency + accountability on these markets
> also nobody denies location is used to personalize pricing
Why would anyone be surprised that location is a factor for a ride service? Location and availability of drivers are basically the two main inputs.
The service wouldn’t make much sense if they didn’t consider location of your start and finish points when pricing rides. This isn’t some nefarious scheme.
There is an undercurrent online, even on HN, of something generally approaching "price discrimination is totally fine so long as it results me paying a lower price." As soon as more "fair" price discrimination schemes end up with the user paying 2, 3, 4x more than others (or than what they've paid previously) it's a great social injustice.
I don't like it either but it's kind of silly to spend most of your professional life trying to help companies maximize revenue then get upset when as a customer you can see that you're helping companies maximize revenue.
> To test the allegation, they used two phones with different battery levels
n=2 is only very slightly better than n=1, and still nothing close to a "debunking". Neither the original claim in the tweet nor the alleged debunking provide anything substantial, and nobody should be claiming anything until they do something at least semi-rigorous (n>10, control group, actual sampling of Uber prices and then statistical analysis).
I couldn't do this, as none of the people in my friend circle have the Uber app, but I'm not in SV - seems like there should be a few HN denizens who could pull off something like this relatively easily and put the question to rest.
> nobody should be claiming anything until they do something at least semi-rigorous (n>10, control group, actual sampling of Uber prices and then statistical analysis)
Couldn't agree more - right now all we have is very weak and inconclusive evidence. If someone is able to confirm this allegation, it would make for a big story and potential lawsuit, especially since Uber has been publicly and categorically denying the claim for at least 5 years.
"taken 10 minutes apart" - a rigorous proof of the claim would need to evaluate pricing at the same moment in time for two different battery percentages.
It isn't beneath them to use dark patterns, they already do that in real life with their wages/practices.
Realistically I think it'd be pretty difficult to test this from the outside and get definitive answers. If they implemented something this scummy they'd surely implement some safeguards to make it harder to prove they're doing it, similar to the VW Diesel Emissions scandal.
This should not be something that alleged without evidence. It would be trivial to take multiple phones with different battery levels and compare the prices. You could also vary the distance between the phones. Furthermore, and I could be wrong on this, I don't recall there being any API to get the battery life remaining.
This is alleged “with evidence”. I understand you want more evidence. It’s alleged and now others can try to simulate.
For what it’s worth, I was able to simulate prices ranges of $50-130 for the same ride with a few friends when we were all standing next to each other. Any new phone shows a low price first, if you search again the price goes up nearly 50%.
So simulating the evidence here might be hard based on other algorithms Uber uses
So if you use rooted Android and modified the OS to always show full battery, you'd always get the best prices...
And what would happen if the API call returned a larger value, say, 10,000%? :)
I struggle to see a legitimate reason for an app to be able to access battery level. It's pretty unlikely that app is going to change behaviour to preserve battery life – that's probably better handled by the OS slowing things down. Better to simply deny the app this information.
Apps can change behaviors to preserve battery life though. Things like disabling prefetching will likely reduce radio time and thus preserve battery.
Delaying maintenance can help too, and that also helps reduce occurances of power failed during complex data operation and the resulting difficulties of resuming from a partial operation when the state may not have been fully persisted at any point, because filesystems and abrupt power loss don't tend to go so well together.
Also things like disabling complex animations, canceling high-CPU background jobs, doing more work on the server side rather than client (trading off latency for battery life), using less complex but less accurate algorithms, etc.
There's a lot of reasons why an app might want to know the battery status, and most of them are for the user's benefit.
First things first: you could develop a custom interface within your app, think of games for example, to show the battery level, or simply an icon when it's low, in a specific way instead of the system's style.
Then, depending on what your app does, you can act on a few things to reduce, or not, battery usage when it gets below some thresholds: reduce GPS precision, remove some animations, limit framerate, make less networking requests, ... It could be limited by the system, but it would probably not be as wise regarding to what is important in the context of one app vs another.
We have numerous cases where we adapt to the battery level so the user can get the best of our apps and not empty their battery while they still need it.
On the one hand, I generally favor allowing users/their apps to get more granular information out of their phone rather than somewhat arbitrarily deciding what they need to know (e.g. there's more GPS detail that iOS could provide but doesn't).
However, in this case, the phone has a battery meter and provides its own low battery warnings. I imagine you could have more sophisticated battery use tracking apps than Apple provides but, in general, not sure why you would legitimately need this info.
Stop with all those permissions already: users are tired of those popups, they don't know or understand what they are asked about, and even when a pro user gets one, it's not possible to really know what will happen/break when it is allowed or denied, so this "solution" is not one, it's a bandaid and it's broken.
My phone disables the camera flash under 15%. I don't see that as anything particularly nefarious. That said I fully support the principle of least privilege.
So Whatsapp tells other participants in a call what your battery life is? By itself, I suppose that's pretty harmless. But it's really none of your business what the status of my phone is in general.
I guess I don't really want an app doing that given it could probably tell participants other things that I don't want them to know, especially if I enabled something like location services.
Permission controls are a bandaid as someone else wrote.
Let's use the location services example.
So some communications app asks me for permission. Do I grant it? Well, as an unsophisticated user, maybe it wants this info to optimize the quality of my connection in some way. So maybe I ought to give it permission?
Instead, maybe it's going to broadcast my coordinates to everyone else on the call which is not OK in general.
There is an api for battery level in both iphones and androids. Evidence: life360 can tell you the battery levels of the phones of people in your circle.
it's not. You need help from an insider to figure out and/or force the accounts that you're using to be opted in to the same A/B tests experiments that surely are used to check the impact of a change like using battery information to determine fares.
Lets not turn hackernews into a fake news dissemination service, and upvote some guy's unproven assertion just because you hate uber. With the millions of rides Uber makes, what are the coincidences of this guy's battery being low and surge price happening at same time?
I doubt this is actually happening due to lack of easily compiled evidence. But, I was curious and looked into price discrimination law a little more. Apparently this isn't illegal.
Stated as a rule, price discrimination becomes unlawful under federal antitrust law only when it threatens to undermine competitive processes in an affected market. I.e. offering a lower price to one party who then goes on to outcompete the other party who you charge a higher price.
In general, there's no reason why it should be. If we were personally negotiating to do some job that I don't particularly want to do--and especially if I have other clients I could work for instead--I will probably charge you more (and you will probably be willing to pay more) if you really need something ASAP.
You need and receive an Uber ride just as fast in the low and high battery level cases. If the claim is true, they would be charging more for the exact same service delivered at the exact same time and pace -- not for earlier or more speedy service. They would be charging more simply because the increased psychological stress means they could get away with it.
I LOLed a little when I realized you really thought for a moment something like this could be illegal in the US.
(I have assumed you are in the US, as everyone else that does not specify the country in a topic that is clearly country specific)
Of course it does. It's just that in my experience living for extended periods in both US and Europe, consumer protection laws are a lot more advanced in Europe. It feels like business regulation in Europe is driven (mostly) by consumer benefit, while in the US is driven by corporate benefit (which a lot of US voters believe it equates to consumer benefit; you know jobs, etc)
Also the US tends to leave a lot more to the market and companies policies are more defined by the legal consequences more than the laws that prevent them to operate in certain manners in the first place.
It would be far too easy for people to confirm that it was happening, and the PR hit - even to Uber - would be way more damaging than any financial benefit they would get from implementing this.
Wow,I needed to take an Uber yesterday, and my battery slowly went from 10% to 5%. When it went to 5% the price doubled, allegedly saying that there was a surge in rides in my area. It wasn't rush hour, and there was no visible surgery as far as I could tell.
The generalized “company flaw” is thinking one is smarter than customers. That’s always incorrect because customers form a large distribution that will always be dissatisfied and want what is better, driving innovation.
But this is a cultural flaw that so many companies make. It’s good to be smarter than others. Life is a competitive resource fight sometimes. But you have to turn off that mindset with customers 100% to be successful in the long-term with them. The customer in the limit is always right. And it’s so much harder to regain customer trust once it’s lost.
Seems plausible, slimy, but also allegedly debunked and not a controlled test (at least show two phones w/different battery levels side-by-side with the same ride).
I find a LOT more concerning the allegations regarding racism and charging blacks more -- and down that chain there is a pic of two phones side-by-side with one black and one white hand, and prices of $57.80 vs $32.58 for the same ride [1], along with other mentions of the same phenomena in the thread. Another one [2] showed $48.91 vs $33.67, but only screenshots. Obviously it is possible that this is somehow manipulated, but, this kind of "personalization" is definitely possible.
What is almost more surprising is that Uber evidently doesn't detect these tests and ensure that two quotes to the same destination from the same location at the same minute aren't shown the same price. If they want to "personalize" according to race, battery level, or whatever, you'd think they'd at least be as good as VW at avoiding scrutiny (after all, Uber is supposedly a 'tech' company, vs VW being merely a car company).
They probably check your # of Instagram and Twitter followers too. Or they will, before long. They have to pull out every trick in the book (and make up some more as they go along).
"True accounting of the last quarter has Uber losing 38 cents on every dollar it took in." [0]
In the app recently I saw Uber offering to drive you around on your errands. They're just pivoting until they get lucky buyout-wise or until they find a business they can actually make money in. The founder was the smartest to cash out; read somewhere he's been investing in ghost kitchens.
Uber doesn't have to invent self-driving for it to have a foundational impact on the industry. Uber with their legislative foothold + market share can still potentially dominate a market with an incredibly low cost of labor if self driving is realized. They would just buy a fleet or the company that makes them.
It occurred to me this would be fairly easy to test. However then I realized this user could be in an A-B test, or this could be a feature-flag that they turn off as soon as a tweet blows up.
If somebody wants to prove it though, they should take a video of the whole process, preferably using 1 phone and 1 account, and show it happening 4-5 times in a row within the space of an hour, and that would be definitive proof.
The tweeter's evidence of battery-based price discrimination is a single heavily redacted photo. It could be true but I'm going to hold off getting outraged until I see some actual corroboration. It's not like it's hard to run a series of controlled experiments here.
Regardless of whether price is based on battery life, it has been shown that price is based on who the customer is and the circumstances. Uber is modeling based on willingness to pay. For example, others have shown that on average, a ride back home will cost more than a ride from home to somewhere else. Uber's algorithm has learned that if you are away from home without your car, you may have few other options, and you may be willing to accept an inflated price.
While this does great things for Uber's short term revenue, it will absolutely destroy it long-term. If the drive to the restaurant is only $10, but the ride back is $40, people will remember this and just drive themselves the next time.
No. I have personally seen two people next to each other get very different prices for the same trip though. IMO it's reasonably obvious that Uber is incorporating its estimate of willingness to pay in the moment to determine the price.
What a joke submission. "Some guy said this" now you're acting like you're some news organization? This should have been shut down immediately. Or, maybe it's a chance for us to reflect on what most news stories really are.
How would people feel if this was "designed" by an algorithm? I don't know whether it was, but ML could discover that people pay more when their battery is low just by having that data and doing some hill climbing.
Personally, I struggle to believe this. Multiple companies are accused of it: all deny it. Surely, if this were true -- and endemic in the ride-sharing industry -- we'd see more than just rumours, whether that's acknowledgement from a company or whistleblowing from someone within these companies.
Uber haven't just used flowery language to evade the question, they've denied in the strongest possible terms:
> It is categorically not the case that the battery level on your phone affects the price of a fare
Once the VC money starts to fade, the competition has been bought or crushed, and the IPO cash out has occurred it’s time to extract money from the users.
I wonder if anyone has looked into the relationship between time-of-day and battery level? Could it be the case that phone batteries are typically lower across the board at certain times of day? If most people aren't charging their phone during business hours, perhaps it's likely that batteries are typically lower around rush (surge) hour. A correlation doesn't imply causality sort of thing.
Could this be a case of correlation =/= causation? If we consider independently when batteries may be low, and when prices may be high, it's possible those overlap. For instance, at the end of the day on Friday or Saturday, after being out late at night, the battery is more likely to be low; similarly, Uber prices are probably higher on weekend late-nights with higher demand.
Regardless of the current usage of the battery info on a user's mobile, why shouldn't they leverage it to extract maximum value? This is exactly the kind of dark pattern a la Booking.com ,albeit behind the scenes, that I would expect from any tech company today. Or maybe I read too much doom and gloom headlines on HN nowadays.
The battery level is likely a single feature in a dynamic pricing algorithm, which has discovered an inverse relationship between battery level and price receptiveness on its own.
I use lyft more than uber these days. Its typically cheaper especially at the airport for some reason. If you want competition make sure you're using the competition.
You see stories like this and think that China is correct in coming down on the tech sector. No one is better off when a company pulls stunts like this.
It’s true that a centralized system can make “the right calls” without having to compromise for the purpose of maximizing or minimizing something… that’s the good. On the other hand, they can also make unilateral and very poor decisions (like the cultural revolution).
This sort of dross is not cool, regardless of how superior you are, or feel you are, to the rest of the community.
Moreover, you've been breaking the site guidelines habitually. Therefore I've banned this account. If you don't want to be banned, you're welcome to email hn@ycombinator.com and give us reason to believe that you'll follow the rules in the future. They're here: https://news.ycombinator.com/newsguidelines.html.
They point out that these rideshare apps can detect what your phone's battery % is, and the TOS says they may access that info.
To test the allegation, they used two phones with different battery levels and compared the prices (at the same time, & with same destination), but did not find pricing differences.
They reached out to Uber who made a statement:
> Uber does not use personalized pricing nor is battery life a factor in determining fares. Pricing is optimized for the network as a whole, not any one individual rider or driver. [1]
Here is the reference Uber provided which describes how prices are set, and how other marketplace functionality works. [2]
[1] https://www.wkyc.com/article/news/verify/verify-does-uber-ch...
[2] https://www.uber.com/us/en/marketplace/