Some of you might have read the short story Lena by qntm [0], which is about brain uploading and the effective enslavement of the uploaded once their humanity is hidden behind an API. That story is actually about the gig economy [1].
Of course having a dishonest Uber driver is annoying, but I think that has nothing on getting suspended or fired by an automated system.
> "Lena" presents a lush, capitalist ideal where you are a business, and all of the humanity of your workforce is abstracted away behind an API. Your people, your "employees" or "contractors" or "partners" or whatever you want to call them, cease to be perceptible to you as human. Your workers have no power whatsoever, and you no longer have to think about giving them pensions, healthcare, parental leave, vacation, weekends, evenings, lunch breaks, bathroom breaks... This is extremely realistic. This is already real. In particular, this is the gig economy. For example, if you consider how Uber works: in practical terms, the Uber drivers work for an algorithm, and the algorithm works for the executives who run Uber. There is a kind of airtight bulkhead separating the human drivers from the humans who control the algorithm. The management layer sends the drivers instructions electronically, impersonally, at immense scale, and the drivers respond using a narrow collection of APIs. This bulkhead is intentionally designed to make it impossible for these two sets of humans to actually talk to one another
> Uber temporarily suspended accounts of some drivers who were suspected of fraud, including when its systems concluded drivers had taken unnecessary detours to inflate fares or accepted trips without intending to complete them
Regardless of how Uber did it, these kind of drivers should be removed from the platform. This happened to me more than once. It’s quite frustrating as an example you want to get to airport and a driver has accepted your request, but you can see in the app the vehicle doesn’t move. One time this happened for all requests on Uber when I wanted to go to airport early morning and had to switch to Bolt
But wouldn't you want a human to be involved in this process?
Let's say you run a legitimate SaaS company, and a cloud provider suspends your account due to an allegation of fraud. Wouldn't you prefer a human is involved?
To me that seems pretty reasonable when people's livelihood is at stake.
> But wouldn't you want a human to be involved in this process?
The article says human review was used for the final decisions. The accounts were temporarily suspended during the investigation.
Investigations take time because you need to allow time for the person to respond with their side of the story and then to have human reviewers investigate, which can take multiple rounds.
Having human review is as simple as having one person watch a dashboard of accounts flagged for fraud and then press the suspend-account button after skimming it. I fully expect that’s what they switched to after getting hit with this.
> To me that seems pretty reasonable when people's livelihood is at stake.
What about the riders who are on the receiving end of fraudulent drivers in the best case, abusive or dangerous drivers in the worst case? It is possible to go to far in trying to protect workers for things that impact paying customers. Thinking of it only as big bad companies versus innocent workers misses the real problems.
> What about the riders who are on the receiving end of fraudulent drivers in the best case, abusive or dangerous drivers in the worst case? It is possible to go to far in trying to protect workers for things that impact paying customers.
Not saying they shouldn't be banned. But a human should look into it. Seems reasonable?
I don’t think Uber or any other platform would automatically suspend the account based on one incident. It would have been multiple people complaining about a driver and then got suspended.
Drivers should be able to appeal and in that case a human should manually review before making the suspension permanent. Most probably that’s how it happens right now, but article doesn’t say anything about how the appeals are processed.
I almost lost a connection to a flight once using Bolt. Driver had confirmed, and at the last minute decided not to come. Of course on short notice could not find another driver. A friendly hotel worker drop me off.
Bolt washed their hands, and refereed me to the their customer agreements:
"Bolt says it is in no way responsible for performance of the transport contract between you and the transport provider."
"Their liability section goes further in Clause 6.1 says Bolt does not guarantee the quality or freedom from defects of the transport service. It says losses connected with the transport contract, including accidents, are not Bolt's responsibility."
F#%&uck Bolt, Uber and all the corrupt politicians who allow these innovative disruption "businesses"
I disagree, all those issues you pointed out are there with a normal taxi as well. However, what Uber has done for passengers is that drastically reduce the probability of getting scammed or abused by drivers.
Try getting a taxi in a country as a tourist where uber doesn’t operate and I bet 99% of the time you would get scammed or charged twice or three times as the normal cost of the ride
Thai taxis are notorious for off meter fares, but a savvy tourist can avoid them mostly (flag down a cab, don't take one waiting for you or worse, offering a ride).
I've made it work for me when the legit cabs had a long wait - I offered an off meter fare and I got instant service.
I once calculated the fare using a taxi fare site and the verbal off meter rate was lower than the estimated one. I gave him the higher rate.
Off meter isn't so black or white, that said I always prefer uber or similar regional cabs.
This fine is for the entire EU, the complaint was filed in France but handled by the Dutch because Ubers european headquarters is located in the Netherlands.
So according to the regulator Uber automatically suspended driver accounts that were flagged in decision of fraud or due to low ratings. They also failed to inform drivers about the automated decision making.
Here is the statement from the Dutch data protection authority:
Uber made fully automated decisions regarding drivers. In cases of suspected fraud or excessively low customer ratings, driver accounts were automatically temporarily deactivated or, in the case of persistently low ratings, permanently deactivated. As a result, their income from Uber ceased during the deactivation period.
Uber thereby violated the ban on fully automated decision-making under the General Data Protection Regulation (GDPR). The AP also found that Uber failed to adequately inform drivers about the automated decision-making. Uber has since ceased the violations.
Reference to the law that was broken (GDPR): [2][3]
Article 22.1: The data subject shall have the right not to be subject to a decision based solely on automated processing, including profiling, which produces legal effects concerning him or her or similarly significantly affects him or her.
Article 13.2f: In addition to the information referred to in paragraph 1, the controller shall, at the time when personal data are obtained, provide the data subject with the following further information necessary to ensure fair and transparent processing: [...] the existence of automated decision-making, including profiling, referred to in Article 22(1) and (4) and, at least in those cases, meaningful information about the logic involved, as well as the significance and the envisaged consequences of such processing for the data subject.
We’ve gotta normalize linking to the underlying ruling. Uber says they didn’t automate any permanent suspensions and the regulator says they did; how can anyone interpret this story without more information on how the regulator determined this?
I am curious, if the algorithm provided the recommendation and a person had to say agree or disagree for each would that be against the law?
Uber may want to reconsider operating in the Netherlands- would be surprised if they get this much in profit from this region. Unfortunately for them these laws are EU wide.
If the person was just rubber-stamping the computer decision, then likely yes it would still be against the law. They would likely need to demonstrate that each case was evaluated and considered by a person with proper knowledge and/or training to make such a decision.
If rubber-stamping decisions by a supervising person is a breach of this GDPR rule, then the judiciary should be very thankful that the legislative excluded it from GDPR's scope :D
Likely true, but I was speaking specifically about rubber-stamping an automated computer decision. The law seems to require human judgment to be applied, and just passing a computer decision unchallenged doesn't seem to do that. But I don't live in Europe, so maybe I'm giving it too much credit.
> GDPR rules ban decisions made solely by computer algorithms when they have a significant impact on people’s lives, such as on employment, saying such decisions require meaningful human review and a way to challenge a decision.
The law Uber broke. I can’t believe they found it that difficult to follow something as simple as that.
Which is ridiculous on it's face, imagine claiming that stopping someone's income until you get around to reviewing it doesn't have a significant effect on someone.
The real issue is if uber comes away from this thinking they can do a rubber stamp review, because the point of the law was clearly to stop unjustifiable decisions automatically being made(protecting both parties), as uber will still be on the hook for unjustified suspensions but redress will need to go through the court making it worse for all involved.
They didn’t find it difficult, they found it expensive and figured they could avoid it. In the US, they’d probably have been right; in the EU, apparently not.
> The case against Uber concerns European incidents from 2018 to 2022, stemming initially from a French complaint. It was handled by the Dutch regulator because Uber’s European headquarters are in the Netherlands.
The data subject shall have the right not to be subject to a decision based solely on automated processing, including profiling, which produces legal effects concerning him or her or similarly significantly affects him or her.
Nowhere in article 22 [1] does it restrict the scope to employees.
If Uber processes data, and makes decisions based on that data in a purely automated way that significantly affects people, it is not allowed. That's all there is to it.
I'm so tired of this talking point. Either follow our laws or get the fuck out of our market.
There's no special right for American companies to ignore our rules and when they get fined for breaking them, that's just rule of law.
Maybe there's been enough regulatory capture that big corps can ignore laws on the other side of the pond, but for now, we are still a society of law and justice.
And it’s also incredibly tiring to force legal paradigms of one country over another. Don’t like it? Don’t operate. Please don’t operate and whine how the country is not bending over backwards to your demands.
The internet has enabled a sort of structural/emergent legal colonialism - not centrally planned or directed, but as an emergent property of a global information system and those most economically incentivized to profit off it. The most immediately unpleasant example of this is nuisance streamers who deliberately antagonize people in foreign jurisdictions while leaning on the norms and protections of their own home turf.
Reading the Guardian article - it feels that the decision is weakly motivated. It is not clear exactly what rules Uber broke.
>“Uber has committed serious infringements” by deactivating driver accounts without warning or human involvement, the organization’s deputy chair Monique Verdier said in a statement.
>“From one moment to the next they no longer had any income … A computer should not make decisions on its own that have (such) major consequences.”
If you manage to extract usable info from that quote kudos. To me it looks like regulator going berserk in reading the unholy mess that is GDPR.
GDPR rules ban decisions made solely by computer algorithms when they have a significant impact on people’s lives, such as on employment. Such decisions require meaningful human review and a way to challenge a decision.
But the drivers are not Uber employees. Which is the whole point of uber existing. It is a shitty model from Uber's side - but in this case just attack the model. Outlaw gig economy and be done with it. There are enough protections in employment laws in EU.
Given this kind of ambiguity, it seems like we need some kind of, like, procedure by which it’s decided whether Uber violated the law as written, and if so, what the consequences should be - maybe the sort of thing where we’d get experts in the law together in a room to investigate and debate the merits of each side’s argument and determine which stands up to scrutiny or not. Maybe we could have third disinterested legal expert hear both sides and make a decision. That’s really the only way to clear this whole thing up, I think.
This is not true in all jurisdictions. Some places have been smart enough to see through this transparent legal fiction and declare Uber drivers to be employees, with Uber having all the relevant responsibilities toward them.
This is classic US common law loophole lawyering. That's just not how EU law works. I discussed this in depth with another commenter in this thread a while ago if you want to know more: https://news.ycombinator.com/item?id=48707697
Are you saying that we should allow a Uber to do whatever they want because of scale they are operating? It’s Uber choice to operate at scale. No one forced them to.
This is pure revenue extraction from American tech. The EU now makes more in fines from American companies than it does in taxes from its own tech industry. It is clearly impossible to fill these roles with humans without breaking the economic case for Uber, which is obviously the goal. Europe is so fucked man.
> Sorry how else are they supposed to do it at Uber scale?
The people at Uber are ubersmart, so they can figure it out. If they can't, they don't know how to run their business in line with the law and should be fired.
Figure out how many humans are required to make the decisions, then hire that many humans and have them do it.
I swear this is like a restaurant chain complaining that they shouldn't be required to follow foodservice regulations because of how much money it costs them at scale.
What an unusual situation, a big company that has to manage infraction reports from their employees. I don't know... how does Starbucks, McDonalds, Volkswagen or Carrefour do? Do they have, maybe, managers or a human resources department? I'm just talking out loud.
Just how many cases are there? Surely not actually that great number that some employees can not manually go over the details. Maybe cut the developer and executive pay by say 20% and you can easily cover cost of those employees.
Yeah, I find the notion that human judgement is superior to algorithmic judgement to be pretty suspect… look at Jens Ludwig’s work on judges, it’s not like involving a human improves outcomes somehow
Some of you might have read the short story Lena by qntm [0], which is about brain uploading and the effective enslavement of the uploaded once their humanity is hidden behind an API. That story is actually about the gig economy [1].
Of course having a dishonest Uber driver is annoying, but I think that has nothing on getting suspended or fired by an automated system.
[1] https://qntm.org/mmacevedo
[1] https://qntm.org/uploading
> "Lena" presents a lush, capitalist ideal where you are a business, and all of the humanity of your workforce is abstracted away behind an API. Your people, your "employees" or "contractors" or "partners" or whatever you want to call them, cease to be perceptible to you as human. Your workers have no power whatsoever, and you no longer have to think about giving them pensions, healthcare, parental leave, vacation, weekends, evenings, lunch breaks, bathroom breaks... This is extremely realistic. This is already real. In particular, this is the gig economy. For example, if you consider how Uber works: in practical terms, the Uber drivers work for an algorithm, and the algorithm works for the executives who run Uber. There is a kind of airtight bulkhead separating the human drivers from the humans who control the algorithm. The management layer sends the drivers instructions electronically, impersonally, at immense scale, and the drivers respond using a narrow collection of APIs. This bulkhead is intentionally designed to make it impossible for these two sets of humans to actually talk to one another
> Uber temporarily suspended accounts of some drivers who were suspected of fraud, including when its systems concluded drivers had taken unnecessary detours to inflate fares or accepted trips without intending to complete them
Regardless of how Uber did it, these kind of drivers should be removed from the platform. This happened to me more than once. It’s quite frustrating as an example you want to get to airport and a driver has accepted your request, but you can see in the app the vehicle doesn’t move. One time this happened for all requests on Uber when I wanted to go to airport early morning and had to switch to Bolt
I agree.
But wouldn't you want a human to be involved in this process?
Let's say you run a legitimate SaaS company, and a cloud provider suspends your account due to an allegation of fraud. Wouldn't you prefer a human is involved?
To me that seems pretty reasonable when people's livelihood is at stake.
> But wouldn't you want a human to be involved in this process?
The article says human review was used for the final decisions. The accounts were temporarily suspended during the investigation.
Investigations take time because you need to allow time for the person to respond with their side of the story and then to have human reviewers investigate, which can take multiple rounds.
Having human review is as simple as having one person watch a dashboard of accounts flagged for fraud and then press the suspend-account button after skimming it. I fully expect that’s what they switched to after getting hit with this.
> To me that seems pretty reasonable when people's livelihood is at stake.
What about the riders who are on the receiving end of fraudulent drivers in the best case, abusive or dangerous drivers in the worst case? It is possible to go to far in trying to protect workers for things that impact paying customers. Thinking of it only as big bad companies versus innocent workers misses the real problems.
> What about the riders who are on the receiving end of fraudulent drivers in the best case, abusive or dangerous drivers in the worst case? It is possible to go to far in trying to protect workers for things that impact paying customers.
Not saying they shouldn't be banned. But a human should look into it. Seems reasonable?
You’re still missing it
Humans did do the review.
They paused their ability to take new rides until the human review was complete.
I don’t think Uber or any other platform would automatically suspend the account based on one incident. It would have been multiple people complaining about a driver and then got suspended.
Drivers should be able to appeal and in that case a human should manually review before making the suspension permanent. Most probably that’s how it happens right now, but article doesn’t say anything about how the appeals are processed.
Depends how pervasive the issue. I agree a human in the loop for reasonable redress is appropriate
I almost lost a connection to a flight once using Bolt. Driver had confirmed, and at the last minute decided not to come. Of course on short notice could not find another driver. A friendly hotel worker drop me off.
Bolt washed their hands, and refereed me to the their customer agreements:
"Bolt says it is in no way responsible for performance of the transport contract between you and the transport provider."
"Their liability section goes further in Clause 6.1 says Bolt does not guarantee the quality or freedom from defects of the transport service. It says losses connected with the transport contract, including accidents, are not Bolt's responsibility."
F#%&uck Bolt, Uber and all the corrupt politicians who allow these innovative disruption "businesses"
I disagree, all those issues you pointed out are there with a normal taxi as well. However, what Uber has done for passengers is that drastically reduce the probability of getting scammed or abused by drivers.
Try getting a taxi in a country as a tourist where uber doesn’t operate and I bet 99% of the time you would get scammed or charged twice or three times as the normal cost of the ride
Thai taxis are notorious for off meter fares, but a savvy tourist can avoid them mostly (flag down a cab, don't take one waiting for you or worse, offering a ride).
I've made it work for me when the legit cabs had a long wait - I offered an off meter fare and I got instant service.
I once calculated the fare using a taxi fare site and the verbal off meter rate was lower than the estimated one. I gave him the higher rate.
Off meter isn't so black or white, that said I always prefer uber or similar regional cabs.
This is likely more than Uber's entire Netherlands annual revenue, I think.
This fine is for the entire EU, the complaint was filed in France but handled by the Dutch because Ubers european headquarters is located in the Netherlands.
So according to the regulator Uber automatically suspended driver accounts that were flagged in decision of fraud or due to low ratings. They also failed to inform drivers about the automated decision making.
Here is the statement from the Dutch data protection authority:
Reference to the law that was broken (GDPR): [2][3] [1]: https://www.autoriteitpersoonsgegevens.nl/actueel/uber-krijg...[2]: https://gdpr-info.eu/art-22-gdpr/
We’ve gotta normalize linking to the underlying ruling. Uber says they didn’t automate any permanent suspensions and the regulator says they did; how can anyone interpret this story without more information on how the regulator determined this?
I am curious, if the algorithm provided the recommendation and a person had to say agree or disagree for each would that be against the law?
Uber may want to reconsider operating in the Netherlands- would be surprised if they get this much in profit from this region. Unfortunately for them these laws are EU wide.
If the person was just rubber-stamping the computer decision, then likely yes it would still be against the law. They would likely need to demonstrate that each case was evaluated and considered by a person with proper knowledge and/or training to make such a decision.
If rubber-stamping decisions by a supervising person is a breach of this GDPR rule, then the judiciary should be very thankful that the legislative excluded it from GDPR's scope :D
Likely true, but I was speaking specifically about rubber-stamping an automated computer decision. The law seems to require human judgment to be applied, and just passing a computer decision unchallenged doesn't seem to do that. But I don't live in Europe, so maybe I'm giving it too much credit.
> GDPR rules ban decisions made solely by computer algorithms when they have a significant impact on people’s lives, such as on employment, saying such decisions require meaningful human review and a way to challenge a decision.
The law Uber broke. I can’t believe they found it that difficult to follow something as simple as that.
They claim they didn't break it.
Which is ridiculous on it's face, imagine claiming that stopping someone's income until you get around to reviewing it doesn't have a significant effect on someone.
The real issue is if uber comes away from this thinking they can do a rubber stamp review, because the point of the law was clearly to stop unjustifiable decisions automatically being made(protecting both parties), as uber will still be on the hook for unjustified suspensions but redress will need to go through the court making it worse for all involved.
Do you have a source for that?
They didn’t find it difficult, they found it expensive and figured they could avoid it. In the US, they’d probably have been right; in the EU, apparently not.
> something as simple as that
Dutch Uber drivers aren’t employees [1].
[1] https://nltimes.nl/2026/01/27/uber-drivers-freelancers-emplo...
> The case against Uber concerns European incidents from 2018 to 2022, stemming initially from a French complaint. It was handled by the Dutch regulator because Uber’s European headquarters are in the Netherlands.
That's irrelevant.
Nowhere in article 22 [1] does it restrict the scope to employees.If Uber processes data, and makes decisions based on that data in a purely automated way that significantly affects people, it is not allowed. That's all there is to it.
[1]: https://gdpr-info.eu/art-22-gdpr/
This is a better citation than the original one.
*Netherlands raises $1bn in special taxes from overseas operation.
No, it’s a fine because they broke the law. The EU fortunately takes that seriously.
Uber is breaking Dutch law.
I'm so tired of this talking point. Either follow our laws or get the fuck out of our market.
There's no special right for American companies to ignore our rules and when they get fined for breaking them, that's just rule of law.
Maybe there's been enough regulatory capture that big corps can ignore laws on the other side of the pond, but for now, we are still a society of law and justice.
And it’s also incredibly tiring to force legal paradigms of one country over another. Don’t like it? Don’t operate. Please don’t operate and whine how the country is not bending over backwards to your demands.
The internet has enabled a sort of structural/emergent legal colonialism - not centrally planned or directed, but as an emergent property of a global information system and those most economically incentivized to profit off it. The most immediately unpleasant example of this is nuisance streamers who deliberately antagonize people in foreign jurisdictions while leaning on the norms and protections of their own home turf.
Reading the Guardian article - it feels that the decision is weakly motivated. It is not clear exactly what rules Uber broke.
>“Uber has committed serious infringements” by deactivating driver accounts without warning or human involvement, the organization’s deputy chair Monique Verdier said in a statement.
>“From one moment to the next they no longer had any income … A computer should not make decisions on its own that have (such) major consequences.”
If you manage to extract usable info from that quote kudos. To me it looks like regulator going berserk in reading the unholy mess that is GDPR.
GDPR rules ban decisions made solely by computer algorithms when they have a significant impact on people’s lives, such as on employment. Such decisions require meaningful human review and a way to challenge a decision.
But the drivers are not Uber employees. Which is the whole point of uber existing. It is a shitty model from Uber's side - but in this case just attack the model. Outlaw gig economy and be done with it. There are enough protections in employment laws in EU.
“Such as” employment.
Given this kind of ambiguity, it seems like we need some kind of, like, procedure by which it’s decided whether Uber violated the law as written, and if so, what the consequences should be - maybe the sort of thing where we’d get experts in the law together in a room to investigate and debate the merits of each side’s argument and determine which stands up to scrutiny or not. Maybe we could have third disinterested legal expert hear both sides and make a decision. That’s really the only way to clear this whole thing up, I think.
People's lives, not just employees.
This is not true in all jurisdictions. Some places have been smart enough to see through this transparent legal fiction and declare Uber drivers to be employees, with Uber having all the relevant responsibilities toward them.
This is classic US common law loophole lawyering. That's just not how EU law works. I discussed this in depth with another commenter in this thread a while ago if you want to know more: https://news.ycombinator.com/item?id=48707697
GDPR is not limited to employment matters.
That is the whole point of the GDPR. The spirit counts, not some checkmark rule. That's how the teleological interpretation works.
Sorry how else are they supposed to do it at Uber scale?
Are you saying that we should allow a Uber to do whatever they want because of scale they are operating? It’s Uber choice to operate at scale. No one forced them to.
This is pure revenue extraction from American tech. The EU now makes more in fines from American companies than it does in taxes from its own tech industry. It is clearly impossible to fill these roles with humans without breaking the economic case for Uber, which is obviously the goal. Europe is so fucked man.
American tech companies just have a hard time operating in a legal system that isn’t bought and paid for.
If it's clearly impossible, how come Uber managed to become compliant in 2022?
> The EU now makes more in fines from American companies than it does in taxes from its own tech industry.
That's because American tech doesn't pay its fair share of taxes to the EU compared to its revenue.
Then Uber is fucked. "This is how you do it cheaper" isn't an escape hatch, and we shouldn't accept terrible behavior just because it's cheaper.
Disclaimer: I briefly worked for Uber and quit in disgust
> Sorry how else are they supposed to do it at Uber scale?
The people at Uber are ubersmart, so they can figure it out. If they can't, they don't know how to run their business in line with the law and should be fired.
Figure out how many humans are required to make the decisions, then hire that many humans and have them do it.
I swear this is like a restaurant chain complaining that they shouldn't be required to follow foodservice regulations because of how much money it costs them at scale.
Is following the law optional if you're able to break it at scale?
These laws are effectively retroactive wrt Uber, and are protectionist nonsense anyway. Good luck.
How are they retroactive? Did Uber not have reasonable time to comply? They are a young company, how can they be too calcified to react?
What an unusual situation, a big company that has to manage infraction reports from their employees. I don't know... how does Starbucks, McDonalds, Volkswagen or Carrefour do? Do they have, maybe, managers or a human resources department? I'm just talking out loud.
Just how many cases are there? Surely not actually that great number that some employees can not manually go over the details. Maybe cut the developer and executive pay by say 20% and you can easily cover cost of those employees.
According to Uber it wasn't a lot of drivers. They should be able to easily have a human review each case.
If they cant figure that out, then they shouldnt do it at all. Duh.
Anyway, if they're so large and special then they can hire a bunch of people. Large scale should make it easier.
Hire people?
Well they can reduce driver pay and hire additional lawyers; the EU way.
Yeah, I find the notion that human judgement is superior to algorithmic judgement to be pretty suspect… look at Jens Ludwig’s work on judges, it’s not like involving a human improves outcomes somehow