To put it in perspective, the average income in Kenya is around $1.25 an hour and tbh (as a Kenyan) I really can't see it as a bad thing. A lot of people here live in abject poverty, they live in situations you can't really begin to imagine. So any sort of help coming our way is good.
I don't think your comment makes it any better. With all respect but given the HDI of Kenya, I don't think your average Kenyan worker is that versed in English, has a laptop and an Internet connection. $2/hour for a company like OpenAI (which can, most likely, afford more) is pure modern slavery.
It's not slavery though, since nobody is forced to take the job. This clearly was something Kenyan workers were happy to have as an option, otherwise they wouldn't have taken the job. If everyone did what OpenAI did and moved work like this to Kenya, salary would quickly go up. If few companies moved their work there, comp will stay low. It seems the low price is one of the main attractors right now. So shaming companies into paying more, likely will just move the work elsewhere and leave workers there with $0/hr.
Why is that an OpenAI problem? It’s something the Kenyan government needs to solve. I fail to see why every company needs to subscribe to an SJW mindset.
They paid decent wages by Kenyan standards it seems. They did not force, exploit or abuse the workers.
Poverty and inequality are terrible. OpenAI should be ashamed that they just made it marginally better for some folks in Kenya instead of solving this massive problem in its entirety that has nothing to do with their company. /s
Yes, some money is better than none. That is exactly why it is better. They are literally doing something whereas you are probably doing nothing (or much less than what they are doing).
The workers have made clear they prefer OpenAI's offer.
We could tell the workers that we know what's best for them. Sitting here in my rich country with my six-figured job, I'm supposed to tell the workers they're wrong, they should have taken my offer instead?
It's a terrible incentive that helping people a little bit will get you criticism while helping not at all earns you no criticism. In fact, as non-helpers we get to sit and sanctimoniously criticize the people actually doing something.
That’s some BS that we were fed about competitiveness. I used to buy that BS at some point. But these companies (and the people having them) have a choice and they do have enough money.
One of the companies I’m following now is Oxide (https://oxide.computer/careers). They pay everyone the same salary ($200k) regardless of position. It’s a bit extreme but am following them precisely for that. It’d be interesting to see how Oxide fares down the road.
It’s funny that some years ago that software developers were complaining that they were getting paid less because of politics and because people who talked controlled the businesses and the money. It’s funny because now that the table turned, most of them are doing the same thing.
Probably everyone with roles that aren't part of the core business is outsourced. Am argument could be made that outsourcing is actually worth for the workers than earning less than highly skilled labor, but being a FTE and maybe get opportunity to grow.
Nothing funny about it, it’s called leverage, and realistically it’s not always a bad thing. The note up above about it not being slavery because it’s voluntary is really important. If someone is entering into an agreement willingly, that’s pretty good evidence at least (if not proof) that they expect or perceive it to be a mutually beneficial transaction. This can be the case regardless of some 3rd party’s (and especially a rich westerner’s) gut reaction that the wage is crazy low.
> If someone is entering into an agreement willingly, that’s pretty good evidence at least (if not proof) that they expect or perceive it to be a mutually beneficial transaction.
Respectfully, the existance, and the need, for minimum wage puts lie to this assumption. People need money to survive, and will take whatever job they can when they don't have one, even if it not for enough pay.
I'm not claiming this is the case here, but that this argument is badly flawed.
Not really? In a world with no minimum wage, working for sub-living wage is better than zero. That’s still beneficial compared to the alternative. So the analogous thing here, which I think is true, is that if you think it’s crazy people are accepting a job for $2, you ought to start looking for solutions amongst the jobs paying $1.99 and below.
That’s what minimum wage does. It eliminates the alternatives below the line, not above it.
> That’s what minimum wage does. It eliminates the alternatives below the line, not above it.
You should go back and look at why minimum wage was established in the first place - so people could work and still have lives outside of work.
And frankly, the "option" of starving out on the street is not a real alternative we should even be including in our discussions in 2023. It smacks of treating poor folks as some subhuman species who has to earn their right to live from us.
Err, right. “The line” I’m referring to is below the wage itself, not below the poverty line or below the level required to live (the minimum wage being well below both of those in much of the country today).
No one is discussing that alternative. Not sure who you think is? We’re discussing whether a $2 wage in a country with an average wage of $1.50 is abusive.
My position is simply that 1) a wage in one country appearing low to the standards of a completely different country is not evidence of abusive employment; 2) people accepting those wages when they’re not coerced is evidence that the wages are not abuse, though it could also be evidence of other problems that preclude a better alternative.
It's telling that you had to remove the conditional from that excerpt you've taken. Rhetorically owned me and made zero progress on the actual conversation and problem we're trying to unpack together. You have a good day as well!
They outsource low paying jobs like "building their hardware" to other companies. I bet OpenAI content raters are also managed by a vendor, but even if not, the difference is smoke and mirrors.
> It’d be interesting to see how Oxide fares down the road.
I have been hearing about Oxide since 2019 [1]. 3 years later, it seems they don't yet have paying customers that one can read about. And with a tag line like: "Servers as they should be" I'm inclined to ask: "Servers as they should be [by whom?]".
Is the demand for such ideal servers driven by engineers/sys admins or by businesses (i.e. businesses who are currently being served poorly by existing options in the market)?
Is the demand for such ideal servers enough to make them a sustainable business, especially with the enormous amounts they'll likely be pouring into R&D before they can bring a product to market that lives up to their ideals?
When we started the company, we knew it would be a three year build -- and indeed, our first product is in the final stages of development (i.e. EMC/safety certification). We have been very transparent about our progress along the way[0][1][2][3][4][5][6][7] -- and our software is essentially all open source, so you can follow along there as well.[8][9][10]
If you are asking "does anyone want a rack-scale computer?" the (short) answer is: yes, they do. The on-prem market has been woefully underserved -- and there are plenty of folks who are sick of Dell/HPE/VMware/Cisco, to say nothing of those who are public cloud borne and wondering if they should perhaps own some of their own compute rather than rent it all.
I was just responding to the specific claim, but I think what they could do is pay as if they weren't leveraging borders to decrease wages. I'd rather solve the problem at the root than require such exceptionalism from them though, as that doesn't really seem scalable.
This is a really good troll. I almost tried to debate with a person on the internet over whether getting a data entry job that pays 50% over the national average was slavery. Kudos, sir.
> $2/hour for a company like OpenAI (which can, most likely, afford more) is pure modern slavery.
by using the same argument, you can claim that someone taking home $300k USD a year living in CA is a victim of modern slavery simply just because his/her employer can afford more.
If we're going to expand the term "slavery" to include at-will employment for pay, we should probably repeal the 13th Amendment and replace it with something a little more precise.
> I don't think your average Kenyan worker is that versed in English
Kenya is a multilingual country. Swahili, a Bantu language, and English are widely spoken as lingua francas and serve as the two official languages. English was inherited from colonial rule (see British Kenya).
It does seem a fair observation that having a group of people that aren't 100% well versed in the language and culture of the subject matter might not deliver the best results.
>a company like OpenAI (which can, most likely, afford more)
Hate this argument. As a for-profit company, OpenAI is legally and ethically obligated to minimize unnecessary expenses. It's purpose is to develop AI systems, not to provide people with high-paying jobs.
Half of Kenyans have smartphones, practically every corner of the country has 4G internet. Large parts of Nairobi have broadband connections to the home. It's not nirvana, but it's really not as dire as you are implying.
Not sure I would call this "help". These workers were subjected to the most vile and graphic depictions of sexual abuse content imaginable for next to nothing.
This whole thing makes OpenAI seem evil if you ask me. Just another company exploiting people who are already being exploited. It's depressing.
Sorry but did you read the article? There were some clear examples in which the workers were subjected to this content.
"That month, Sama began pilot work for a separate project for OpenAI: collecting sexual and violent images—some of them illegal under U.S. law—to deliver to OpenAI"
"Sama delivered OpenAI a sample batch of 1,400 images. Some of those images were categorized as “C4”—OpenAI’s internal label denoting child sexual abuse—according to the document. Also included in the batch were “C3” images (including bestiality, rape, and sexual slavery,) and “V3” images depicting graphic detail of death, violence or serious physical injury, according to the billing document."
The point of the OP comment is that it's not "next to nothing". The cost of living differs from place to place and a small amount of USD can be worth a _lot_ in many developing nations.
A couple of years ago I seriously considered joining the federal police department in charge of dealing with internet crime here in Germany, so I've thought quite a bit about this topic.
Basically, it is a job that needs to be done in society, but one that is torturous, and can leave you with long-term or permanent problems. In essence though, it is not fundamentally different in the way a coal miner would jeopardize their physical health, just with mental health instead. This risk/possible damage should be rewarded with a higher wage, and adequate measures should be put in place to minimize the possible damage, eg. in the case of content moderation with access to therapy and only exposing employees to short intervals of traumatic content.
This is of course how things should be, in reality coal miners working environments only reached a decent level through unions and a long fight for better rights. Content moderators are not paid well anywhere either.
My belief: When you have a legitimate choice of your place of employment, and all of the opportunities will let you live your definition of a comfortable life, that's when it's no longer exploitative.
So many times - especially for poor folks - there is no meaningful choice. "I do this or I don't have a place to live." "I hold two jobs so I can feed my child."
> My belief: When you have a legitimate choice of your place of employment, and all of the opportunities will let you live your definition of a comfortable life, that's when it's no longer exploitative.
By that argument, paying 200$ an hour would be much more exploitative. This would be like landing a job paying $6 million a year in the US; it would be insane to quit such an opportunity, especially since every other opportunity is basically poverty (not even in comparison). Following this logic, it's 'graceful' to only pay 2$/h, since that makes them equal to the other opportunities and therefore not exploitative (while still paying reasonably well).
Effectively, it seems like you're calling OpenAI exploitative based on factors they can't change.
> paying 200$ an hour would be much more exploitative
I didn't say that - I only said that it depends on choice. Does the employee have a choice if they have a $60k (average individual income in the US) option and a $6M option? Yes. Are they de-facto forced into taking the $6m one? No.
I know many people who didn't take higher paying jobs, or left such jobs, because they knew the high paying job was going to be miserable.
There was even an article about one such individual just the other day here on HN: Quitting the Rat Race
"I’m currently working at a top tier investment bank as a software engineer. I’m an insignificant cog in a machine that skims the cream from the milk. I’m earning the most money I’ve ever made and yet I’m the least fulfilled I’ve ever been."
Maybe I should have steelmanned your position a bit more. That being said, the grand³parent said:
> This whole thing makes OpenAI seem evil if you ask me. Just another company exploiting people who are already being exploited.
In that context, there is no way for OpenAI not to be evil, since they are (by definition) only one option in the market. In fact, taking your argument to the extreme, there is no way to offer jobs in Kenya as the first company to offer jobs would either be exploitative by paying minimum wage or exploitative by being the only real option. Going from that, paying a higher wage just worsens the situation, as it makes the alternatives even less feasible.
That being said, I do get where you are coming from. But it is not a good point to accuse OpenAI on, as they are making the situation better by offering options at a (for a Kenyan level seemingly reasonable) rate and they really don't have any other option[0].
[0] Except maybe paying Americans a lot of money for the job, but I find it morally hard to argue that they should pay US citizens a lot of money instead of paying Kenyans (comparatively) good money, even leaving aside economical feasibility.
As a war veteran I can confidently say most veterans would probably not be traumatized by work like this. We've all seen much worse and the things that became problematic memories for me had little to do with reading/hearing/seeing the worst humanity has to offer in terms of violence and fucked up shit people do to each other. The stuff that really sticks and eats at you is usually stuff that happened to or close the individual or something that happened as the result of actions taken by the individual or their immediate group.
Probably the most important measure a company could take to prevent lasting harm with this kind of work would be to spread it around a whole lot more than just 36 people. The real risk of long term impact here would probably be with persistent exposure to it all day long. Most people can handle reading or seeing some graphic stuff with proper mental preparation for it but to see nothing but that day and day out would quickly wear you down.
>Is it ever appropriate or ethical to ask humans to voluntarily subject themselves to traumatic experiences in exchange for compensation?
Probably not. Yet people still voluntarily sign up for military service around the world by the millions, and they do so for a bunch of personal, family, idealistic, cultural, and societal reasons that are hard to reduce to a few easy to argue points like a lot of people online try to do with stuff like this.
Personally, I think it's admirable to hold the ideal that we should like to never offer jobs like this, we should also like to never offer jobs that involve going to war, cleaning up hazardous materials, dealing with explosives, working around heavy fast-moving machines, cleaning sewers, or a myriad of other terrible experiences either; but we're probably not in a position to make those better choices just yet. Until then, people are willing to do these things for a dozen different reasons per person, only two of which are the pay and support they get from the employer.
It's kind of insulting to tell someone that you know better than they do what's good for them.
A Kenyan has told you that this job pays substantially better than average and that they wish more companies would make such jobs available, and your response is "no, you're wrong, this job pays too little". On what basis do you make that claim? What makes you better qualified than OP to comment on conditions in Kenya?
Fair point. But this is also the exact problem with allowing a small number of individuals to accrue massive wealth by arbitraging labor costs like this. It doesn't matter how much philanthropy Bill Gates engages in, he doesn't understand the needs of the poor better than they do.
Instead of having this elaborate, inefficient system of funnelling money to first world billionaires and then having them (maybe) send some of it back to the developing world, wouldn't it be better if these workers were just paid better in the first place?
In theory, yes, but at some wage level it wouldn't make sense to choose Kenya over India. At another threshold it wouldn't make sense to choose India over the US.
There is overhead to sending jobs to poor, unstable countries in timezones far from headquarters. If we insist that everyone, worldwide should be paid the same wages we are in the US, what incentive do companies have for not just hiring locally and avoiding those costs? If we step it back and say "okay, you should at least pay what you would in India", then why not just hire in India, which is a far more predictable environment than Kenya?
The job pays too little based on the fact that it can leave the "employees" traumatized and scarred for life.
But it does not take away from the fact that it can negatively affect the employees in an adverse way long term, and takes advantage of poor people to give them an objectively bad deal based on an information asymmetry where the person in question might not know that they have to read graphic descriptions of bestiality and pedophilia.
Scarred for life by reading/labeling text?
There is obviously a big difference between labelling pictures or video and text. I would be open to seeing an actually study, but my prior would be that text must be harmless. I would certainly be open to reading any text, especially if the context is that the text is for training purposes.
This is an argument that the job shouldn't exist at all. That's an okay argument to make (I personally lean that way), but has no relevance to whether $2/hour would be worth it to a Kenyan.
I think people can make that choice as long as it's informed. The company should also be providing mental health support as part of the job. Someone has to do it. There aren't automated systems that can do a remotely decent job at moderation yet. I know you don't think it's okay for trained systems or social networks to distribute traumatizing material.
Just because a group of people is accustomed to poor working conditions doesn't make it any more right or ethical.
You could apply your logic to sweatshop scenarios where the people in those countries are just happy to have work, even if the work pays unfair wages, requires unreasonable hours, uses child labor, and provides no benefits to the workers. But hey... the disenfranchised are just happy to have a paycheck right?
I'm not a labor wage specialist so your guess is as good as mine. Do you think $1.32/hr. is fair? Are you of the impression after reading the article that the worker's wellbeing was taken seriously and the pay was set at a fair amount considering the kind of work they were doing? I wasn't.
If you're not a wage specialist and my guess is as good as yours, why were you so quick to dismiss someone who actually lives in the country you're opining on?
It's clear you're one who enjoys a circular argument so I'll just leave it at this for you. I don't need to know the exact right amount of money these people should be paid to know that $1.32 per hour for looking at child porn and violent imagery is too little, especially without the requisite mental health resources available. If you are so sure this is a fair situation, maybe for your next job you'll accept minimum wage pay doing similar work.
It isn't pictures, it is text only. I think there is a huge difference. I had to police content for Twicsy (a search engine with 10 billion Twitter images indexed) and I have seen some very bad stuff. There is a huge difference.
This is just false, if you read the article Sama also collected explicit, illegal images on behalf of OpenAI -- this was the reason the contract was cancelled.
If for the local market $1.32-$2/hour take-home is good compared to alternatives (of which I have no idea, but local claims seem to support that, listing comparable rates but for gross not take-home), then yes, it's fair, and it would harm the workers if we'd prohibit that, because they would have to take a worse local job.
It’s a necessary stepping stone on the path to better working conditions and wages. I think people forget what the early days of the Industrial Revolution looked like in our countries.
Can you get there without that? Likely not.
What you’re suggesting is to actually keep them poor for their own good. It’s a nonsensical and counterproductive argument that your making.
Not exactly. I'm suggesting that work like this be paid at a fair rate and mental health precautions are considered and taken seriously.
You can chop logic on this all day long if you like, but the point is, this work is terrible and damaging, and that's why we farm it out to countries like Kenya where the people there don't have a choice.
The work is better than the options people have there, otherwise they wouldn’t do it. Don’t try to spin it as a negative thing for them. They don’t see it that way.
Kenyans are able to win this business because they can do this job at a competitive price. If Kenyans would require more, they would not get this (relatively good compared to their alternatives) job, it would go someplace else and Kenyans would lose out.
> Just because a group of people is accustomed to poor working conditions doesn't make it any more right or ethical
I think this kind of moral puritanism is an enemy of real social progress. Maybe we shouldn't or can't expect some supreme, pure state of "ethical", maybe all we can or should expect is "better".
what's insulting is that you assume one kenyan speaks for all of them. even worse, that the random kenyan not being paid $2/hr by openai is qualified to speak on their behalf. on what basis do you accept the first statement. if another kenyan claims the opposite would be then become some vacillating organism between two (or more) positions?
You also have to realise most lot of people here are not sheltered. We've experienced post election violence first hand in 2008, tribes killing other tribes, we also have a culture of burning petty thieves alive. We experience violence first hand. I don't think violent text is going to affect anyone the way you think it is.
A lot of people aren't sheltered, but it doesn't mean they should be required to do traumatic work for little pay. Also, it was more than looking at violent text. Just because the people there already face hardships that may be greater than looking at graphic text and imagery doesn't mean the world should just pile on because their daily lives are already bad. That just makes a bad situation even worse.
Sounds like you haven't read the article either. It was more than just reading text but the text was traumatic too:
"One Sama worker tasked with reading and labeling text for OpenAI told TIME he suffered from recurring visions after reading a graphic description of a man having sex with a dog in the presence of a young child. “That was torture,” he said. “You will read a number of statements like that all through the week. By the time it gets to Friday, you are disturbed from thinking through that picture.” The work’s traumatic nature eventually led Sama to cancel all its work for OpenAI in February 2022, eight months earlier than planned."
> We experience violence first hand. I don't think violent text is going to affect anyone the way you think it is.
I'll be glad when this kind of "suck it up buttercup" bullshit is gone from our world.
Yeah, violence hardens people. Most often to the point where they're one light tap away from shattering. PTSD is a real thing, and just because folks in Africa aren't being diagnosed with it doesn't mean witnesses to this violence don't have it.
This kind of attitude - that since we've seen some shit we're immune to it all - is just a badly misplaced sense of pride.
I'm not taking any pride in that statement. I'm just trying to highlight how because of the way our culture is, these sorts of texts are relatively tame. I'd actually like for us to get to a place where violence is not ingrained in our society.
It’s a job that is safer and pays better than the alternatives. Don’t go imposing your view of the world on others and thinking you know what’s best for other adults. You won’t like it if I come into your life and do that, even if I were right.
In this case is text. There’s no graphic depictions of anything. There could be foul, abusive, or racist language. But that’s much less difficult to deal with.
You sound like a few of the other commenters here who haven't read the article. Go and look at some of my other comments where I've quoted excerpts for others who couldn't be bothered to read before commenting.
Compare this to being a developer in Canada and working for a US company. Coworkers in the US make more as a base salary and pay less taxes than Canadians do even though the actual cost of living is not different. The argument there is its not about the value of the work delivered, its the cost of competing in a market. Say your average dev for that position makes 140k USD in the US, and in Canada, the average dev for that position makes 100k CAD (74.5k USD), most will pay that dev a modest amount over 100k CAD to attract the talent and compete in the local market, say 120k CAD, which is just below 90k USD. Is it abuse to pay them less than others who do the same work in the US? Most who do the work in Canada are probably in agreement that it would be nice to make the same amount, because who doesn't want more money, but in reality, its still a good salary in comparison to most other jobs locally.
Its not exploitation if you're paying people higher than their local cost of living and higher than other local jobs. If you're just appalled about people being paid an amount of money that to YOU, based on your cost of living isn't fair because you'd want to make more, then be appalled at capitalism as a whole and how much work is being hired outside of company's originating countries. The production of almost everything globally is outsourced to locales where the cost to compete is lower than in the originating country.
More than anything else these companies call themselves high tech. They act as if they are doing revolutionary things. But under the surface its just armies of cheap labour cleaning up ever growing mountain ranges of shit.
I also think it is probably good. Similarly when I was born in 1980's in Poland my father's salary was roughly 0.1$/hrs. My first job in 1999 paid me less than 2$/hrs.
If any American company would be allowed to hire pepole in Poland in 1980 for 1$/hr people would pay bribes to get this job even if it was a humiliating, damaging, life-shortening physical labor.
Why Poland got richer after 1990 is partly that we allowed western companies to buy cheap polish labor. The labor was good enough that the companies started competing for it and long story short now the salaries aren't 5% of western but depending on the sector maybe 20%-50%.
Yeah, in eastern EU salaries are not as low as they used to be, anyone with 3-4 years of experience now expects around 3.5-4k€ a month, which in my experience is not that far off of UK salaries.
Not so in Greece. Salaries are way lower than 3,5-4k per month. Doctors, for example, in public hospitals get around 2k per month and the consultants maybe around 2,5-3k.
I believe that when ricardobayes said "anyone with 3-4 years of experience" they really meant "anyone in our industry". In my part of Eastern Europe, 2.5k EUR is a very good salary, but software devs earn way more.
To me it looks like before. Translates to up to 2.6k€ take-home-pay provided you don't take advantage of any tax breaks and you're on a permanent employment contract.
That being said young people(less than 26yo) are either exempt from the first tax bracket(~20k €) or start off as contractors who pay lower taxes and contributions.
I was a student in the former Soviet Union in early 1990s when the old system collapsed, taking with it stipends that were sufficient for school cafeteria food. After that I occasionally worked for 30 cents per hour doing somewhat-synchronous translation for preachers who came from the US to convert the masses.
While the money sounded laughable (the US minimum wage at the time was $4.25/hr) and some employers got borderline uncomfortably weird, working 4-5 evenings for a week allowed me to feed myself for a couple of months and focus on my studies. I never considered myself to be taken advantage of. Just my 2c.
In early 1990s most of the population was declared atheists, but because atheism was one of the pillars of the rejected Soviet rule it was being reexamined by the people. So there was a lot of orgs of all kinds who flew in to preach and increase their denomination base.
The fact that to the outside Russia looks orthodox Christian today mostly reflects last 20 years of the state efforts to build an official religion, not some deep predisposition of the population. My 2c.
As soon as I started reading the article I realized that "SamaSource" in this case is the same company that was outsourced to help moderate content on Facebook. IMO the point of contention isn't that $2/hr isn't a livable wage in Kenya, is that $2/hr isn't enough when you're talking about looking at vile content for an entire shift. Granted this is ChatGPT so at worst you're tagging vile text rather than tagging vile images and videos.
This raises more of a modal/philosophical question for me which is: should we be outsourcing some of the most vile parts of social media? Is it fair to go to someone in a developing country and offer them a wage lightly above average in exchange for having to look at this content for hours? I have no answers to these questions myself but it's a question I ask when social media companies outsource the content moderation.
could the average income in kenya be around $1.25 because companies decide to pay abysmally? or the people don't merit any higher? would openai be offending the kenyans if they paid them, say, $5/hour? i think this is the question you should grapple with. openai will be selling this tech around the world, and i wonder if these $2/hr workers can afford it at all. would they, or their loved ones, be able to enjoy produce of their labor?
Look for "Nash equilibrium". Game theory makes it really hard for all companies in a free market to conspire to keep wages down. Can they conspire to a small degree? Certainly, probably disguised as a cultural or traditional bias.
Employers don't pay $5 per hour because they wouldn't make a profit either.
That's a theory that only models certain market conditions. In practice there is an information asymmetry (contrary to the presuppositions of the Nash Equilibrium) and often the employer side has a small number of companies so coordination is easier. Check out this settlement from the 2000s in Silicon Valley[1]. This involves highly desired, well educated employees so imagine how it is in other fields.
In my own experience I've had several prospective employers at conferences say that they were interested in poaching me but mentioned that they were wary to start a poaching war. They knew the CEO of my company and were conscious of the fact that he would likely make an effort to respond in kind. The Nash Equilibrium is an interesting concept but reality is complex and messy.
Take your pick. Even just in the tech sector in silicon valley there are many examples of collusion. Mind you this involves some of the most desirable employees on the planet.
But back to the central point the basic nash equilibrium is in the spherical cow realm of models as they relate to actual reality. There are some interesting additions to it involving time series and asymmetrical information to try and model real economic data.
> Employers don't pay $5 per hour because they wouldn't make a profit either.
Nope, they might be able to make profit at $5 or $10/hr, but there it doesn't make sense to pay more than you have to, that's why Chinese devices are significantly cheaper than their western counterparts, they can still compete if they raise the prices but they can allocate cheap labor in china that allows them to become a strong competitor
We're talking about African economies. Those businesses don't have margins that would allow a five times higher wage while still making profits.
Big tech companies, maybe. But even those.
Once you raise prices the demand shrinks which also leads to smaller profits. There's no free lunch. If employers could make more money by paying higher wages, they generally would.
Free enough. Especially in large enough countries the market is free enough. There is no such thing as a completely free market. But in general, and in most places, no, most employers can't conspire to dump wages. Even if it looks like conspiracy, it may be something else, like a bidding war on the supply side.
I've heard arguments in the past of paying "acceptable" US wages to people in other countries would destabilize the other country more than it would help. I'm not an economist, so I don't know how valid those arguments are. I have always thought they were more scare tactics used in favor of being able to pay those low wages for as long as they could get away with it. I could see though how paying a small-ish number of people wages that dwarfs the larger number of people could cause a bit of turmoil.
to me, it goes back to the argument about remote workers should be paid less than the in office worker. or workers living in cheaper areas should be paid less than those living in expensive areas. to me, the salary should be paying for the work being done and that's it. if the same level of work is being done by both employees, it shouldn't matter if one is in Kenya and the other is in the US. the value of the work is what should be getting compensated. again though, i'm no economist.
> the salary should be paying for the work being done and that's it.
I agree with you. The issue is quantifying the essential cost of work. As companies sometimes have no idea how & where they lose or earn money, they tend to rely on external factors to select a "justified" level of wages: average industry wages, location, local laws, whether you need to commute etc.
>I've heard arguments in the past of paying "acceptable" US wages to people in other countries would destabilize the other country more than it would help.
Which honestly is a bullshit argument. The only destabilizing thing here is it would make the person in the higher-paying profession more valuable to society and more able to make economic impact where it might not normally. In this case, a rising tide does lift all boats...because you're giving more economic purchasing power to more people. Will this kill some businesses or disrupt the status quo? Yeah. Will it help more people than it hurts? Maybe, probably, who knows? But I personally see no problem with paying people relative to their output, no matter where in the world they are.
Having been to Kenya myself and worked with Kenyan developers for several years, most of whom were quite talented, I see absolutely no problem with paying them their value relative to the work being done. I have no problem with high skilled Kenyans making much more than the local average because they have a skill that is in demand all around the world.
>the salary should be paying for the work being done and that's it
100% agree. If a business derives $X profit from a laborer and agrees to pay a given % of $X to the laborer as compensation, it should not matter if the laborer is in a high cost or low cost of living area - the business still makes the same amount in profit, and shouldn't get to say "hey yeah we made $10mm off your application, but we're going to only pay you 5% of our profits as compensation because you live in Bangladesh and the cost of living is lower, so be happy with what we're giving you" when if you did the exact same labor and lived in, say, LA or NYC, they'd give you 10%. Just bullshit IMO.
The civil engineer’s experience rings true from what I’ve seen, but there’s some exception for developers and people of other high skilled professions because they are few relative to the population…paying local developers the same as foreign won’t destabilize but it may for manual labor almost everyone can do.
Sure, but doesn't that mean their absence will have a disproportionate effect on the industries that rely on them? If you're a small 15-person dev company, and now there's a multinational company that's gobbling up developers at 4x the going rate, you're not going to be able to compete.
This is one of those things in which it's better to sudden upsets (even if they're "positive"), might have negative consequences.
If you're paying +25% over the going rate, you're going to attract a range of people. Someone that designs software for reactor controls might not care (or be compensated enough anyway).
However, if you're offering 200% over, you're guaranteed to hoover up the top talent from strategic industries, and that might end up being a net negative in terms of the damage caused.
Your small local company does not own your workers. If the multinational values your workers more than you do, then it is good for everyone except you that they work for the multinational.
But, conversely, the multinational values your workers because it knows how to use them for business opportunities elsewhere to make money. But you're in a position to find many of those same opportunities. Which now means that your local economy is not just getting the profit of having the workers do so well, but of the fact that you're keeping the profit margin that otherwise would have gone to the multi-national!
Free trade on average makes everyone richer. (Observation originally due to Ricardo.) That means that it brings both opportunities and risks. And the opportunities usually exceed the risks.
At least in the case of Honduras, more people clearing paths for transmission lines accelerates the development of infrastructure that the locals can use. On the other hand, OpenAI moderation doesn't do anything for Kenyans. You can see the effect of this is countries that heavily rely on tourism like Thailand and Bali. The smartest and hardest working people end up serving foreigners rather than their domestic population. Places that aren't catered towards tourists are completely neglected.
"On the other hand, OpenAI moderation doesn't do anything for Kenyans."
Have you worked at OpenAI in Kenya?
Because you're speaking in absolutes-- as though you're an authority on the topic.
Let's use basic logic: Working at a technology company by its nature provides:
- Increased experience with Computers
- Increased experience with Data
- Increased experience with Business Processes
- Increased experience with AI Technology & AI Business
- Increased experience with Potential business ideas for startups due to exposure to the business processes of the company and the potential therein for improvement
The GP told you: they've been to Kenya themselves and worked with Kenyan developers for years. Please read the whole comment before you fly off the handle at someone. This is unfortunately increasingly common on HN.
i wonder if they would have ever hit saturation here. you can only hire so many people. i would compare it to the discomfort that is felt by private transport operators in the wake of a public transport scheme. when the government of ghana introduced public buses that charged less than half the regular fare, they shook the market. but the public buses could only carry so much--many, due to their circumstances, were served by the private operators. it remains so to this day. in fact, private operators have flourished.
likewise, mining companies launching in ghana have usually paid higher than the incumbent. it could be strategic as it allows them to hire highly skilled workers without the cost of pre-training. here too, a saturation is reached, and things stabilize, with people doing the same job but receiving different compensations. we at devcongress (https://devcongress.org) have worked hard to ensure that local companies pay top-dollar for tech talent. we enforce this through our job boards. still, there are companies out there paying $500/mo and they receive a flood of applicants. but these are my experiences. makes me believe that after the market has been excited, it will definitely dissipate the energy, and return to previous equilibrium (or achieve equilibrium at a higher orbit). i honestly believe that an equilibrium will be reached.
In addition to having lived (for a short time) in Kenya, worked for a company based out of Nairobi for nearly 5 years, and worked with Kenyans that whole time (in other words, I've seen this stuff firsthand in Kenya, which is the whole point of the article, and my comment)...I have also lived in, worked in, and studied in other developing countries, and seen firsthand what happens when high-paying companies swoop in and change the economic landscape.
You know what actually happens? Positive economic change. Isn't the whole point of HN (and YCombinator) to disrupt the status quo, and promote growth? Destabilizing a corrupted, top heavy, status quo where more people now have a chance to grow in life rather than live in abject poverty is a good thing. Nobody cried for the Taxi companies when Uber & Lyft came along, because they destroyed a parasitic, rent-seeking system (Taxi medallions) and the experience for the end-user (both driver and rider) was significantly improved. Same logic applies here.
>The result:
--> He was threatened with harm by bosses of local companies.
Why? He took all their workers.
Yet there's a flipside here that you're completely missing. Those same workers had a significant improvement in their lives, most likely able to accumulate what for them was generational wealth, or enough to give their kids a shot at a real chance in life. Those people had enough to stimulate their local economy much more than anyone beforehand. THAT is what drives positive change in 3rd world or developing countries. Keeping things the way they are economically with no disruption in the 3rd world keeps people poor and unable to grow which leads to bigger problems down the road. Giving them a real chance in life by paying them more gives them and their families a larger ability to make real change in their countries and communities. That should be celebrated.
Put yourself in the worker's shoes - if someone offered to 10x your wages, what would you do with the extra wealth that you now posses?
Having also lived in that part of the world (Mexico, which isn't the quite same as Honduras but analogous enough) , I understand how things sometimes play out in cases like this. The local bosses in your example were probably doing quite well for themselves and keeping a fat cut of the output of their laborers who were paid pennies, and not happy that their game was now being played by a bigger fish than they could fry. So, they acted like a cartel and tried to threaten the new king in town. Did some of the local companies in your example die? Yeah, probably. But nobody is entitled to stay in business....that's sort of the whole point of globalized capitalism - you compete, and you can win or lose, sometimes through your own fault, or sometimes through no fault of your own. Sometimes someone will steal your lunch, and there's not a damn thing you can do about it. That's the reality of it.
Your assumption is so short sighted that it's hard to explain how wrong you are.
Let's not think about the noble but the practical: did the man whose life was threatened continue to stay in the country and pay those workers, or did he leave or drop wages? Did people make decisions about where to work based on the idea of long-term compensation? What amount of people left an otherwise stable job to take this high-paying gig that disappeared? What is the net impact on these people _especially_ if the distortion was prolonged and did put other companies out of business?
This is not a 'compassionate' thing to do, it's short-termism that messes with other people's lives. Moreso depending on the function of how large the distortion was and the relative purchasing power of the job. An example of how this works is AWS jobs in other regions -- they pay significantly higher than base rate in many developing countries, but it is both a pedigree and leaves people in a position to get another white collar job that's fine later on. However, the same is not true for something like a gig-work cleaning agency paying 3 times market rate. I have the feeling based on the source that your experience might be closer to the former, which might leave you with a different take than what's happening here.
My assumption may or may not be short sighted, but is a reflection of the fact that I've spent a while in developing countries and understand that people want nothing more than to provide for their families - yet their opportunities to do so are significantly limited, more than anyone in the developed world could understand....and there are both internal and external forces in these countries that want nothing more than to preserve the status quo, which is that they get a huge cut and the workers get jack-fucking-shit to the point where $2-3 USD an hour is a massive windfall.
Maybe the guy in the example did leave the country, maybe he didn't. But either way, those workers were given a massive windfall for a short time, and I'm willing to bet that some of those workers took that money, invested in their families to the point where their children were able to get some form of education, and those children didn't have to lead the same impoverished lives their parents did. They were able to use that money as a stimulus for real growth. That counts.
I've lived in the developing world too. There's a reason why locals who actually want their country to succeed generally hate this. Nobody says "paying well" is a bad thing, but you have to realize that well is relative.
As someone who has been poor, the idea that the massive windfall is better is so frustrating. It's not. At least where I lived, people could not explain to you how a bank worked. People often did not save money, and if someone found out you had money, you were immediately asked for 'loans'. Beyond that you have to think about _lifetime earnings_. If you hire all of the lumberjacks for your slick handmade-in-africa table business, what happens to the SYSTEM of these people when other businesses go out of business and liquidate their assets then you leave? It's not as simple as some new business popping up and employing people.
Ah yes, the timeless principle of I have not, therefore others must have not. Is that the basic crux of the argument you are making? Or did I misunderstand?
If I may point out, the argument (upholding the system) appears to be structured around justifying why others must have not by saying that all non-high-paying lumberjack-employing businesses will go out of business, and then this high-paying business will leave. Is there any reason why this high-paying business would leave? It sounds from your statement that this is guaranteed, for whatever reason. And if it does leave (because obviously no business is guaranteed for all time) is there reason to believe another external business would not step in to fulfill that demand for handmade-in-africa table business? Or that the newly unemployed lumberjacks (but flush with cash relative to local conditions from their high-paying salaries) will sit around and twiddle their thumbs for all time instead of starting their own lumberjack-related business?
Who got hurt by "destabilizing the labor economy"? The bosses who were perfectly fine resorting to threats and extortion? It certainly wasn't the workers.
It is fine to recognize that people who wield economic power might be upset when a new economic power arrives in town, but to describe the new guy who is trying to do the right thing as somehow immoral is classist nonsense.
It is like when you give away mosquito nets and put all the local makers out of business. It is worse in the long run if you tank the job market then leave.
If you plan to continue to give out mosquito nets for 30 years then it's fine to put the locals out of business, better that they spend their time building a comparative advantage in some other industry.
If you are doing a 1-and-done kind of project, maybe you have a point. You would have to weigh the cost of the disruption of the local industry against the benefit to the workers though. If you pay what would normally be 5 years worth of wages for a year of work, it's fine to set the local industry back 3 years. The workers still come out ahead and can invest their wages into rebuilding industry.
Right but businesses famously operate on 3-month intervals, not 30-year intervals. No business can actually make a 30-year promise like that.
I also don't think the calculus is as simple as "setting it back 3 years", all you would be doing is creating power vacuums (instability), whereas you ideally need constant upwards pressure. It would be like injecting cash via lotteries, it won't actually help in the long run, just create further instability and wealth inequality.
Your argument tells me never invest in any business you create because you will likely crater it. Everyone is hurt by overpaying for labor because you distort the labor market in the exact same way monopolies distort markets above marginal cost, causing dead weight loss for society. A small subset of the harms:The company doing the hiring earns less and so can compensate shareholders less, other companies in that labor market are harmed by being artificially priced out of labor,all labor in said labor market is harmed by the price distortions the overpaid labor often causes, if the demand for labor at the artificially high price isn't durable and long lasting everyone is harmed when it goes away (and it's not durable if it isn't driven by fundamentals and is instead driven by fads for overpaying for labor and virtue signalling about it in the first world because fads change and more rational heads will eventually prevail).
The entire rest of the local economy. If you eat up a significant portion of the local labor poor with very high wages, who is left to do all of the normal local jobs? Working in retail, food production, sanitation, construction, education, etc.
What happens to prices when a significant portion of the workforce suddenly has tons more money? What happens to the people not lucky enough to have the new high paying job? What happens to the local economy when you leave?
When you pay locally crazy high wages you seriously distort the economy for everyone, it doesn’t just magically snap to a new normal where everyone is happy and free, you can very easily create huge wealth disparities that make situations for many people way worse off, and the people who “win” that you don’t employ were generally already winning.
Just pouring money into a highly impoverished place can be very harmful. “Above average” is good, “ridiculously above average” and you have to start considering your effects across the economy.
Seems pretty clear to me (pretty much an identity) that labor paying more is good for laborers but bad for their employers. Not sure how to argue with that.
> Having been to Kenya myself and worked with Kenyan developers for several years, most of whom were quite talented, I see absolutely no problem with paying them their value relative to the work being done. I have no problem with high skilled Kenyans making much more than the local average because they have a skill that is in demand all around the world.
We're not talking about developers though, we're talking about people with good enough English and internet connection to be able to tag or flag strings of text and images. That's not highly skilled work even in a country where native English fluency and internet connections aren't as universal as the US. It's just that the cost of living differences mean that the "get paid just enough to make rent" money that unskilled labour is worth in the US is "more than the local doctors" money in Kenya.
>That's not highly skilled work even in a country where native English fluency and internet connections aren't as universal as the US
Just as a point to that - English is one of 2 official languages in Kenya, and is the national language of business and education. Practically everyone speaks it well enough to get by.
It often requires extra overhead to work with someone in Bangladesh. Tax implications, legal complexities, potential communication & timezone challenges.
Also companies often can't tell how much profit they will make and can almost never tell how much profit each employee is contributing towards. IP, machinery, business processes, brand value, all contribute to the company profits. Some companies with very significant profits can even exist with zero employees. This idea that a laborer deserves some percentage of business profits is just fundamentally flawed.
> If a business derives $X profit from a laborer and agrees to pay a given % of $X to the laborer as compensation, it should not matter if the laborer is in a high cost or low cost of living area
That would be true if that were deal the business had with their employees. However, you made up the percentage thing by observation and are working from the assumption that because you can represent a wage as a percentage of a profit margin, that means the business agreed to share a fixed percentage of profit with the employee. This is incorrect reasoning.
Businesses generally don't contract with their employees to share a percentage of revenue or profits, with the narrow exception of commissions for people in sales. With regard to developers and almost all other employees, the contract both parties agreed to is almost always a trade of specific amounts of time for specific amounts of money and possibly some flexible additional benefits. And to the employee's benefit, the amount of pay does not flex based on whether the company is even producing a profit from the sale of the the products of their labor, only how many hours they put in. Caveat all manner of fuckery in labor relations, of course. Which is a giant caveat, I agree in advance.
However the fundamental agreement is still time for money without the employee having to worry whether the products of their labor can be sold for more than their hourly wage, that's the business's problem to deal with. The employee expects their paycheck to show up on time all the time, or the deal is off.
With that sort of agreement, it is absolutely expected and reasonable that the business keeps any profit from selling the product of the labor after the agreed upon fees for said labor have been paid. That was the deal.
---
If you'd like a different deal for most people, as I would love to see more of in the world economic future, then most people need to change over from needing a job to needing some form of income to support their long term future growth as they live currently off of savings.
Instead of expecting someone else to ensure there is a buffer of money to pay you so you can pay your expenses, you will have to keep that buffer yourself and be okay with the fact that sometimes you will get paid a lot and sometimes not so much, but you will then be getting that percentage of the profit you wanted and it will be totally reasonable to expect the same percentage regardless of what the cost of living is.
You also won't have to worry about layoffs as much because the company's costs now also scale directly with it's profits, so there is no need to cut head count when things get really bad and the company's buffers run low.
But again, that also means you will be responsible for keeping a big enough buffer yourself to deal with the fact that your income fluctuates with both short an long term market swings. Otherwise all you have done is trade the annoyance at seeing the company profit massively from your fix-time-cost labor for the annoyance of being broke even though you have a job whenever the market takes a shit.
Most people still don't want to deal with that, even in super wealthy countries like the US. In fact most people in wealthy countries actually scale their own living expenses in step with any increases in their income such that they can be living paycheck-to-paycheck even at $200k+/yr.
So the balance of things present day is that most people take full-time or even part-time contracts as employees and hope they don't get fired at the wrong time in life.
This is possible to change though, for yourself and for others if you like. So if you think your current deal is bullshit, feel free to change the deal.
When Mansa Musa toured Africa he built new mosques and paid the local workers in gold. Sounds great - skilled labour should get paid well, and he was paying very well. He ended up destabilizing (some people go as far as saying "wrecking") the Egyptian economy for decades. Basically put 22 tonnes of gold into a gold-based economy and suddenly it was deemed less valuable.
The basic argument is a little bit different. There is a huge population in Kenya living, or in some places, failing to live for under $2/day. Most of these people are independent -- for example, street-side sellers in an urban slum, or living in a village far from Nairobi.
The key question is how we improve lives of these people.
It is difficult for me to overstate the impact a 5-fold improvement in income has on people's lives. I've seen initiatives which train people to do tasks like basic data entry, bricklaying, or otherwise, taking $2/day workers to be $10/day workers. Programs like these are typically a few months long. They don't train people to have the equivalent of even a middle school diploma, and they won't be competitive with even the least educated Western workers.
Still, breaking generational poverty is a long process, and that basic increased stability is a first step.
If we shut off employment paths following those programs by giving bad PR to anyone who employs people for $2/hour, we've doomed a key pathway for hundreds of millions of people to escape abject poverty.
Westerners tend to group levels 1-3 together. However, the gap in quality-of-life from level 1 to level 2 is much greater than the rest of the stack. Level 1, life basically sucks. Hunger, lack of basic life-saving medicine, and early death.
Levels 2-4, I've lived at (for at least a few months), and it's okay once you get used to it.
It takes surprisingly little to bring people from 1 to 2, and these sorts of jobs are one way to do that.
As a footnote: People often confuse financial stress with simply being poor, since in the US, those correlate almost completely. Financial stress sucks. Most people worldwide living at levels 2-3 don't have high levels of financial stress. A much more typical situation is a village, where no one has a lot of stuff, but people own the land and their homes. Financial stress sucks. I'd rather own a home at Level 3, with savings, a stable family, income, etc. than live paycheck-to-paycheck at Level 4.
Thanks for sharing, the Gapminder level stratification (and Dollar Street photos) was very eye-opening and a humbling reminder of how privileged I am as a software engineer.
A few years back, Bill Gates bought every university graduate who wanted one a digital copy of the book Factfulness. It's like the insight from that stratification, over and over, in book-length format. It's really quite excellent.
One of the interesting trends you'll see is we've gone from /the vast majority/ of the world at Level 1 really not all that long ago to around a billion people today (probably less when as I write this).
The world is getting a lot better very quickly for a lot of people. At the same time, solving the problem is no longer intractable.
I will also mention: "Privilege" is complex, and life isn't all about money. Starting around level 2 or 3, "stuff" usually isn't the biggest problem in life. You get used to the inconvenience of needing to boil running water or having to ride a bike instead of a car pretty quickly. You don't experience it as inconvenience at all if you've never had it (as most of humanity hasn't for most of time).
On the other hand, humans are social creatures. Many places where people live at level 2 or 3 have communities and families of a type you've never experienced if you grew up in the US. The US is a very lonely place, and there's a mental health crisis in the US that would be completely foreign in many places at level 2 or 3.
The US also experiences much greater financial stress. If you're in a poor village, but you grow your food and own your home, you stress out a lot less about money than if you're making a 6-figure income, but worried about college debt and making your mortgage if you have a short-term job loss, divorce, or medical crisis, as is common in the US.
OpenAI is operating rationally as a corporation, and choosing labor, legally, in a cheaper country - if they were to pay US wages, why not hire American workers? The entire point is because it's cheaper. There's a different world where OpenAI chose to hire Americans for this job, but it's simply cheaper for them not to. The loss of American jobs is a whole other discussion, but it's the very economics of the situation that drove them to do that in the first place.
that were first thrust into poverty to start with by a different system of exploitation.
I have to admire the beauty of this argument as it has occurred over time - let's exploit them using force and make them poor (enriching ourselves), then use the poverty to justify paying them less (enriching ourselves), using the argument that we are slowly lifting these billions out of poverty (enriching our souls).
> I've heard arguments in the past of paying "acceptable" US wages to people in other countries would destabilize the other country more than it would help.
Interesting. If true, that sounds like a good argument for capping salaries. Would the US be a little less chaotic if we set a maximum wage?
Boy, that's a brave thing to suggest on a forum of people with aspirations of getting cushy jobs at FAANGs to get those uncapped wages. It appears it's going over as well as a lead balloon too.
cap income at say 300 m annually then gradually lower it. Get all those mentally ill people out of decision making. Make exceptions for truly mad skills.
Around 50 m have some construction where further work is paid with a salary for life or until the company goes bankrupt.
Lovely! Have an upvote, such as it is!! It is rare to see such a complete demolition of an argument or the real-time validation of that demolition through the rules-for-thee-but-not-for-me reaction to it!!!
central banks' main task is to keep economies from im- or exploding. if there's suddenly a big influx of money they have to raise interest rates to balance that out.
development economics is not about the wages, it's more about market access. the problem is that many developed economies (looking at you USA) have protectionist tariffs which make it hard for developing ones to gain market share.
Would you have a problem if a bunch of billionaires moved into your neighborhood? What do you think would happen to the prices around you? Why do you think people in, say, Lisbon, are so dismayed at the amount of "digital nomads" moving into their cities at the same time as their housing prices rise wayyy above their average wage?
> to me, it goes back to the argument about remote workers should be paid less than the in office worker.
Can you make a stronger link here? Because I'm not seeing the connection other than wages are involved.
Phil Knight mentioned this in Shoe Dog. I don't remember the details but he was paid a visit by some officials and they told him he cannot pay factory workers more than they can pay their doctors.
I don't know about "destabilizing" per se, but whenever people try to set prices significantly above or below market, there tend to be consequences. If OpenAI paid $5/hr when people are more than willing to accept $1.5/hr, demand for those jobs would far outstrip supply, leading to intense competition for those positions. Situations like this often lead to corrupt practices where the bureaucrats/middlemen who find themselves in a position to determine who get these jobs end up capturing a significant portion of the profits, either directly or indirectly.
why $5 then? seriously asking... it also sounds low. in reality companies pays employees (even an american company with an american employee) the bare minimum they can pay them and still have them do the job. the "could the employee afford it" test is a little strange - an astronaut probably can't buy a spaceship, I can't afford some enterprises software that i wrote myself, etc etc.
i suspect that when we look at what caused a raise in salaries at a certain place it is rarely companies from abroad paying more, but simply the local economy developing, causing people to ask for more (which often meant that international companies stopped hiring there, e.g. china).
I can't about Kenya, but this wage translates to around $240 per month (1.5 per hour, 40 hours a week, 4 weeks a month).
This is the average rate the government pays secondary school teachers in Ghana, after 2-3 years of experience. It's almost 2x the rate that private schools pay teachers.
With this wage, no allowances, a single person can live independently in an apartment, and eat out every single night of the week if they choose. It's a fine and decent wage for what amounts to work only requiring a good pair of eyes and basic literacy.
>* a single person can live independently in an apartment, and eat out every single night of the week if they choose.*
And that's a good thing, because you're increasing the spending power someone who will keep the money velocity high has, which by definition helps the local economy. They are spending that money, which will move about and cause growth.
Spending all that money also increases demand on goods, raising prices for everybody else eventually. The people who serve the direct needs of those who get the first dibs on the money are likely still better off, but the further away in that chain, the less favorable the balance. At some point, it may well become negative. So I don't think there's a universal answer here that applies to all such situations in general - one has to crunch the numbers to make any meaningful conclusions on the overall economic effect.
There is some inelasticity to some classes of goods, so you don’t immediately have to worry about upturning an entire nations company because a foreign company decided to pay middle class wages to a few dozen people.
Now for instance, the class of goods we were talking about is restaurant food. This may be a pure substitute for home cooked food, leading to no increase in food demand and money simply being spread to the service workers. Or it could be that they eat different food (e.g chicken imports rather than local fish), which also wouldn’t affect the balance of demand/supply for the poorest and neediest.
Worth and value are inherently subjective. Price is an attempt to assign an objective value to something subjective. It kinda/sorta works, but it also kinda doesn't.
Let's imagine there is no minimum wage in America. Now let's say that workers in America demand $10/hour and African workers ask for $2/hour for the exact same job. What's the correct value for that labor? It's a non-sensical question.
You could say it's the minimum value at $2. But then some other worker asks for $1. Did you make a mistake? Was it not actually worth $2 the whole time?
If they are inherently subjective, why are you claiming we're not paid what we're worth? This statement is meaningless, isn't it?
> It kinda/sorta works, but it also kinda doesn't
In what way it doesn't work? Clearly it has more utility than other attempts at measuring value. I can agree it's not perfect (nothing is), but that's very different from not working.
> let's say that workers in America demand $10/hour and African workers ask for $2/hour for the exact same job. What's the correct value for that labor?
The correct value is whatever someone is paying for that labor on a free market.
> You could say it's the minimum value at $2. But then some other worker asks for $1. Did you make a mistake?
The only mistake is treating value as inherent and static.
I don't understand this argument. Ask for what you think you are (your time is) worth. If you aren't receiving that, you are (your time is) probably worth less
Your argument doesn't take into account power dynamics. When negotiating price, each side comes to the table with bargaining power. If bargaining power is sufficiently uneven then you really can't claim they'll come to a fair agreement.
When an amazon warehouse worker has to pee in bottles to meet a quota do you think the worker and amazon came to a fair agreement? Or perhaps amazon had more bargaining power.
As any other place, the salary is low because people don't have money, or high because people have money.
The more people go there looking to hire somebody, the quicker it gets pushed up. But salaries do basically never change suddenly anywhere, because it's the local people that pay most of them.
(And, no, this is not fair. And locals should push for higher pay.)
At a rate of $5 or more they may be competing with higher skilled jobs in that country (I haven't checked, but four times the average wage...). Doctors and teachers would quit their jobs to earn this much.
That is what's happening in Cuba where highly skilled people with good English skills rather work in tourism than as doctors, engineers or teachers. Not a good long-term situation.
If they were going to pay $5/hour, why wouldn't they do the same work in India, Vietnam, or some other country with wages in that neighborhood, but with a more stable political environment, better infrastructure, and more experienced and educated workforce? Choosing Kenya at that price point would essentially be charity, and might end up being costly charity (because of the extra costs associated with doing business in Kenya vs more established markets).
And then Kenya doesn't gain the benefit of the $2/hr wages (which, as mentioned, is already above average), plus the benefits in knowledge & experience, improved infrastructure, and secondary jobs, as well as the possibility of similar jobs in the future.
China started out competing almost exclusively on wages (like Japan, Korea, and Taiwan before them), and now the US and EU worry about Chinese advanced industries (eg. semiconductors) surpassing their own, the same way Taiwan did with chips and Japan did with cars.
Competing via low wages is, unfortunately, a necessary first step to growing an economy.
Paying too high can often lead to more extortion. There are people who worked as domestic helpers from Philippines or Indonesia at other countries. They are paid about US$500 a month, which is 5-10x of average salary in their home country. The bad news is that you would need take huge loan, with multiple family members as guaranteers, to pay for the “training fees”.
it's common in the US to use US-dollars (and the magnitude expectations around) them when you want to imply somebody is making a tiny salary in another country with a completely different economy.
How much does rice cost? Beans? Other subsistence food? If it's all cheaper than the US, then not pointing that out (or showing how a wage compares to the average for a country), it's misleading journalism.
Minimum wage should never be a benchmark; if someone is paid minimum wage, their employer would pay them less if they were allowed to.
Which makes $2 a RELATIVELY better wage, however for the work - basically content filtering through the worst of the internet - there should be a higher compensation and psychological help available.
In this case, if Facebook and other online services had it available, they would have been better off reusing what they already developed for their side of content filtering. I want to believe 99% of bad content is already automatically filtered out, and anything that comes through and flagged by content moderators fed back into the systems that automate the process.
> The data labelers employed by Sama on behalf of OpenAI were paid a take-home wage of between around $1.32 and $2 per hour depending on seniority and performance.
When one talks about the average income, does that average only consider people that work? Otherwise, in countries with a high rate of unemployment, I would imagine that a single salary might have to feed the whole extended family, so that one needs to have a salary that is much higher than then average income to be sustainable.
Is something deemed bad in one place acceptable in another if it isn't as bad as the alternatives? I could give numerous examples where this is not the case. So assuming we will be consistent in what we consider acceptable or wrong, what then is the deciding factor?
includes a section about Google hiring Americans to do vaguely similar work, at $15/hour. I say "vaguely" because I don't think my cousin Missy had much racist or sexual content to deal with; at least that wasn't the whole point of the job like the Sama example.
Some of you may have heard this story before. The big company uses a small company to be insulated from the legal complications of hiring. That figured into the story: Google claimed they couldn't fix an obvious inequity, and HR told them to try harder. Missy got paid.
(In case you're wondering, that's not a fake name. I showed this to her beforehand, and she said it was fine.)
But are the companies / people in Kenya paying that $1.25 average because they can’t afford to pay more? OpenAI can afford to pay more, but they’re just taking advantage of a less robust economy. Just because someone benefits from something doesn’t mean they aren’t being taken advantage of.
The point of bringing jobs to Kenya is that they can pay less. If you don't see that as a benefit, why should I, a business, go to the trouble of working with a Kenyan company rather than just do it locally?
I could pay a Kenyan $17 an hour, or I could pay a Kansan instead and get US government benefits. The same applies at every payment level. Your logic fails the moment you realize that not every person/worker on earth is equal.
Insisting on this sort of articles just goes to show how sheltered these offended people are. I live in the third world and I know people who are thankful to Coca Cola for their "$2 wages."
Do you think then, that they deserve to be mentally scarred without any support and paid less than the minimum wage for a receptionist, on the basis that they are not "equal" to us?
Why do you think Kenya's economy is any less robust? The per-capita GDP has increased fivefold in the past two decades. It's a fraction of the income level of the US, but it's a robust, dynamic, and growing economy.
(And I think most of what you wrote is based on similar stereotypes)
I can think of few better outcomes than someone "taking advantage of" 200,000 people living in Kibera slum in Nairobi by giving them $10/day desk jobs. Or some of the poorer rural populations. Or the hundreds of thousands some of the refugee camps.
This is a nice sentiment, but I think the answer is way more nuanced. You can't just roll into a developing economy and pay way over market without also disrupting the local economy and the people that live there. Imagine some similar situation in America, where for some reason, an international business comes in and pays 10-30X the market rate as similar businesses in the area, for the same product. The new jobs become highly (and potentially dangerously) desirable, other similar business go under because they can't keep up with the wage growth, etc.
To remain stable, economic growth must be slow and steady. The alternative is you simply don't go to Kenya, rather, you go somewhere else, and Kenyans get $0/hour.
It’s not a sentiment. It’s the definition of taking advantage of. Of course there are various arguments for justifications, but it doesn’t change the situation.
I also didn’t suggest anything like paying way over market.
So then what? If paying $2/hour is taking advantage, and paying over market isn't part of your solution...what is the solution? Not hire people in Kenya?
Also...yes it is a sentiment (a view of or attitude toward a situation or event; an opinion). Our opinions about what constitutes "taking advantage" are different. Saying otherwise doesn't make your argument more compelling.
The concern about danger really does not follow your hypothetical situation. Anyway, our view of capitalism would suggest that the failure of companies that use labor less efficiently is a net good.
I think the question we are all considering is why OpenAI behaves differently overseas than they might when trying to poach a smart engineer from a competitor in the Us.
OpenAI did pay more. They paid 12.50. The employees doing the labeling not even getting half of that is awful and they were definitely taken advantage of.
> OpenAI can afford to pay more, but they’re just taking advantage of a less robust economy.
This is a bad take. Globalization (what you call "taking advantage") raised a billion people out of abject poverty in the last few decades.
When a company pays more than the prevailing wage, they're doing good and deserve kudos. Expecting economic deals with less developed countries to be driven by charity (which is what you're advocating for) rather than mutual benefit is A) paternalistic and demeaning, and B) cannot possibly be the basis of sustainable economic gains for developing countries.
> What kind of business model is one that relies on exploitation of extreme poverty in order to succeed?
that's a very good question, but asking it as if it is a problem of OpenAI and not a problem of literally all of our economy sounds a little strange to me.
Created an account just to disagree with this. I’m sorry but this is a totally backwards attitude.
We need to get to a place where the labor of human beings has a globally defined fair minimum wage. It’s outrageous to me seeing the conditions at Foxconn where young college age children spend decades sitting on a production line doing skilled labor for a fraction of the US minimum wage in order to feed their families, where if they were compensated fairly they could spend that extra time getting an education to escape poverty. There’s no way to slice that that isn’t exploitation and the arguments supporting this won’t age well. It’s right next to slavery, it’s a high offense and we need to call it out bigtime and that we don’t really defines a low point in our moral stance. Because tomorrow we might just find that we are no longer at the top of the food chain and that generation are going to remember how we exploited their poverty to steal their time, keep them poor and keep them subservient.
You are right - equal pay for equal work is an easy concept to understand - it is basic fairness. Instead, Africans are played off against local labor groups and it leads to structural imbalances like women still being paid only 82 cents on the $. And folks go around ridiculing equal-pay advocates as "social justice warriors in SF or NY" or "self-anointed do-gooders"!
> We need to get to a place where the labor of human beings has a globally defined fair minimum wage
That is just economic nonsense. Labor productivity of economy depends on utilization of capital and ideas in that economy and it is pretty hard to increase. Average wage per hour in that economy cannot be higher than average productivity per hour. If you set minimal wage higher than its productivity, then you just destroyed almost all legal employment and forced everyone to grey economy or subsistence agriculture.
It’s not nonsense to pay a fair wage. It’s basic human decency. Actually, it’s a sacred obligation. If someone gives you some of their time - literally a slice of their living experience, to help you build your wealth then you need to compensate them fairly. It’s not complicated, not at all. I’d like a detailed explanation as to why you think a young skilled worker in China deserves less than an American performing the same task. Explain it to me like you would face to face with a Foxconn worker over coffee after a 10 hour shift on the production line. Explain that economic nonsense like I’m five
This salary is higher than the minimum in Brazil. Of course, without labor rights. They would probably also get thousands of employees in Brazil if they needed to, as well as in several other countries.
I myself do work with mediation of content for several social networks, getting a little more than that. The only problem is that the monthly hours are irregular, there are very few hours.
If someone has an offer of 4 dollars an hour for 200 hours a week without any benefits, I accept it immediately. And that's half the US salary, as I understand it.
Have you been on the internet in the last 3 months?
We’ve had schools changing policies, entire products being created to detect if a piece of work is AI-generated, countless articles on the impressiveness of the technology, even more articles arguing about some aspect of it.
And this was all on HN.
ChatGPT has been one of the most immediately picked up and heavily used and discussed technologies in the last few years. And it has permeated into the regular zeitgeist in a way that hasn’t happened in even longer. I don’t understand how you could even remotely say otherwise.
I don't think the GP is speaking to Kenyans wearing Indian-manufactured clothes -- I think that they're pointing out the hypocrisy in Time readers (and its authors) who live in highly-developed countries, wearing clothing made by low-wage workers, who are just looking for any reason to criticize OpenAI, without applying that same criticism to themselves.
Ah yeah, the classic “in order to criticize a thing, you must first criticize and root out every possible worse thing.”
People have no way of knowing where the things they buy come from or how to source things ethically unless people write articles like this and push for higher standards when they see an opportunity.
It’s one thing to protect genuinely well-meaning people who can’t meet a purity standard, but OpenAI has a track record of dishonesty (starting with their name).
> People have no way of knowing where the things they buy come from
I don't know about other places and maybe I'm an outlier but I check everything for where it's produced / manufactured. Most if not all items in the US are required to have where the item was produced/manufactured. Some may say "Manufactured in the US with globally sourced material" makes it harder to pin down where each of the components originated from. Clothes and other items the country of origin is on the tag.
As a resident Indian, as far as I know (a) children have a guaranteed right to education which keeps them in school until 14 (b) children in fabric sweatshops was never an Indian thing. Children in brick kilns and beedi factories rolling beedis because of their nimble fingers used to be a thing.
As of 2021, India had an estimated 160 million child laborers. Children in textile factories is absolutely a thing. Look up Sumangali scheme for just one example.
Wiki says there are 200 million odd child labourers globally. As per the Indian census of 2011, the number of child labourers in India is way lower than your 160 M figure.
India’s census figures are mostly reliable, unlikely to be off by an order of magnitude.
As per this link, there are 1.8 billion children globally (i.e. below 14). India has roughly 250 Million children so around 14% of the world population of children.
There are roughly 250 million child laborers globally of which 11 million are Indian children so around 4.4% of the global child labor population is Indian.
I.e., despite having 14% of the world's kids, only 4.4% are child laborers. I'd say India is moving in the right direction when it comes to child welfare.
So, having abandoned your points A and B above, your new point is that being that there are countries with worse child labour problems, India is moving in the right direction?
None of the quoted stats or linked articles support the assertion that things are improving. The only trend that I can see noted is that Indian child labourers are moving to the cities to do manufacturing work and abandoning rural/agriculture work which on the face of it doesn't seem like a fully positive development development all.
It's another instance of the same problem -- that wealthy western society and it's trappings rely on ensuring that there is a set of poor people who will jump at the opportunity to be exploited
Extend this logic to remote work and your salary should scale strictly with where you're living. Living in Ohio but working for a bay area startup? Expect to get paid a "good salary locally". Why not, right?
That is the case in the vast majority of tech companies, though. They (almost) all have something like a "regional adjustment", "location based" or other classification for salaries in different areas. Some are more extreme than others. The company I work for (it's a relatively well-known tech company) certainly has it, and I know Google and other FAANGs do, too.
Point is that when you interview in SF and then leverage remote work to go wherever you want they cannot police where you are and determine the appropriate salary.
OpenAI went shopping for Kenyan labor and those who did the job were not American and never interviewed in SF
Headline implies there’s something negative here, but I’m not seeing it. Was OpenAI supposed to hire high-skill workers in developed countries to do this work?
It's just the dichotomy of using $2/hour labor to build a product that makes you obscenely wealthy.
As another poster said- in Kenya a job is a job and they still paid above average. Money going to Kenya is a good thing. They are doing the work so it must be worth it.
This also seems like a much better service job than the horrible scarring moderation jobs at facebook or factory work for apple. Heck maybe even better than most Amazon US jobs at the warehouses
> Almost more interesting in that article I found that Sama, the Kenyan company, was also asked to collect CP and brutal/gore images for OpenAI. After delivering 1400 images, Sama cancelled the contract as this was even too much for them. OpenAI then talked about "miscommunication" and they actually did not really want those CP images. Well, they made its own category for it and asked Sama to collect images of several other categories, but somehow it was lost in communication that for one of theses categories, no images should be collected. Because they are illegal. OpenAI swears they never opened the images they received (and paid for).
> The San Francisco–based A.I. startup is projecting $1 billion in revenue by 2024, as was first reported by Reuters and independently verified by Fortune. OpenAI only generated approximately $35 million in revenue in 2022, sources say, and the company is heavily loss-making at the moment.
"But these days the human cost is just another cost" - this is where I think you, "key_stroker", and the GP, "brookst", might agree on something, in my opinion.
If someone agrees with your premise that "these days the human cost is just another cost", then it seems logical that a newspaper editor/author will shape the headline about economic suffering, instead of mental suffering.
By adding more information to the headline about "$2 per hour", it might make the discussion more confusing by much of the comments discussing the economic part. BUT, a clickbait-y title might increase the reach of the article.
At the very least you'd hope that they'd provide proper mental health support for people doing that; from the article that appears not to have happened.
That's a serious problem and seems worth an article no matter what country the people are in or what typical wages are in that country. Funny that "$2/hour" struck Time as more newsworthy than "exposed workers to horrific material without training or counseling."
Didn't headline just state the fact objectively, and any negative implication comes from the mind of the reader? If not, how could the headline be less editorialized (in order to reduce the negative implication)?
"OpenAI hired Kenyan company to make ChatGPT less toxic"
"Kenyan workers helped make OpenAI's ChatGPT less toxic"
"OpenAI paid Kenyan workers minimum wage to work on ChatGPT"
Many different things are objectively true, but of course they mention the pay, because they know that to most of their "Western" readers it sounds outrageous.
The pay is not even close to the most outrageous way to headline this article: it uses "toxic" as a euphemism for, e.g., not generating CSAM, rape, and other sexual violence.
If an AI is generating novel images, I wouldn't call them CSAM even if they fall under a rule that makes them illegal.
If they're not properly novel then it depends on what the source images were, I guess? But it's not like you'd need to train it on anything obscene for it to figure out how to combine bad things with nudity.
IMO, the headline ignores the crux of the article: that OpenAI contracted with this company to collect and label data for "a safety system" that would classify and filter out "child sexual abuse, bestiality, murder, suicide, torture, self harm, and incest." In the course of doing that, workers were (naturally) exposed to source data in those categories from the internet until the contracting company cancelled the contract.
I don't have a strong opinion on this myself, but I am curious how the general sentiment of "that's a good wage for Kenya and this is fine" might change if they were hiring developers there instead.
If quality developers could be hired for $2 per hour, that would be fantastic. Imagine how much progress we could make. The first thing I would do is hire a bunch of people to fix the bugs in my favorite open source software.
In the short term it might make my life harder as I would probably lose my high paying development job. But that's absolutely not a good reason to oppose progress.
That's how it's working, many companies have developers in low-cost areas, and they definitely aren't paying them USA-level salaries, often not even USA-minimum wage.
Developer salaries even in such countries tend to be far better than $2/h, but that's explicitly because western companies like these "were hiring developers there instead", if they weren't, the developer salaries there would match the local market and be even lower.
So long as developers want to be paid by the hour rather than paid for a specific result, I would say this is fine too.
The problem everyone is waffling on here isn't the competitiveness of the wage, it's the gap between the value of what the labor produces and the value offered to produce it.
But you can't escape that gap so long as you're fundamentally trading time for money.
If you want to capture more of the value of your work, you become a contractor and bid firm-fixed or cost-plus contracts instead of labor-hour or time-and-materials.
If you want to capture even more value than that, you work on building business systems and selling a valuable product or service rather than spending your work hours doing the tasks themselves.
Whether you're in Kenya or the US and whether you're doing undesirable or highly desirable work, it is not a moral failing of the business you are working for that they did not also take the time to ensure that you could have become a contractor if you wanted to or that they did not ensure that your economic situation allows sufficient market opportunities for you to quit and form a business if you wanted to.
Almost more interesting in that article I found that Sama, the Kenyan company, was also asked to collect CP and brutal/gore images for OpenAI. After delivering 1400 images, Sama cancelled the contract as this was even too much for them. OpenAI then talked about "miscommunication" and they actually did not really want those CP images. Well, they made its own category for it and asked Sama to collect images of several other categories, but somehow it was lost in communication that for one of theses categories, no images should be collected. Because they are illegal. OpenAI swears they never opened the images they received (and paid for).
Directing someone to commit a crime is still a crime. OpenAI most likely has criminal liability in this instance and the FBI should open an investigation if they haven't already.
You have it backwards; OpenAI was the one that made the request. OpenAI claimed that the Kenyan company misunderstood the request. Your parent comment is claiming that OpenAI is criminally liable for making the request.
Crime is not the appropriate word but there's clearly a well documented history of abusing conditions toward data labeling workers [1] [2] [3]
For those looking for an AI data labeling service also tries to help workers along the way, here's a plug to the company I started "dataprep.be" [4]
We have a small preference for working workers with special needs (deaf, mute and employees with small handicaps). Public subsidies for these type of workers help our case in the EU as well as contraints for some public institutions in the EU to hire more handicap workers.
With clients more sensitive to costs, we also work with remote data labelers from developing countries. We help putting checks in place to limit forced and child labor. We pay 5% extra so they have time to learn high demand tech skills. (Using Khan Academy and free access to a normally 250$/year Datacamp subscription)
Happy to work with the HN crowd or just receive feedback and mentoring!
(My email is in my profile)
The fact that you refer to workers in "developing countries" as being leveraged for clients "more sensitive to costs" should tip you off that what you're doing is exploitative and dehumanising.
Is there a source for this claim? This would be life-ending for OpenAI if true.
edit: This is mentioned in the original article. From TFA:
>Sama delivered OpenAI a sample batch of 1,400 images. Some of those images were categorized as “C4”—OpenAI’s internal label denoting child sexual abuse—according to the document. Also included in the batch were “C3” images (including bestiality, rape, and sexual slavery,) and “V3” images depicting graphic detail of death, violence or serious physical injury, according to the billing document. OpenAI paid Sama a total of $787.50 for collecting the images, the document shows.
> Within weeks, Sama had canceled all its work for OpenAI—eight months earlier than agreed in the contracts. The outsourcing company said in a statement that its agreement to collect images for OpenAI did not include any reference to illegal content, and it was only after the work had begun that OpenAI sent “additional instructions” referring to “some illegal categories.” “The East Africa team raised concerns to our executives right away. Sama immediately ended the image classification pilot and gave notice that we would cancel all remaining [projects] with OpenAI,” a Sama spokesperson said. “The individuals working with the client did not vet the request through the proper channels. After a review of the situation, individuals were terminated and new sales vetting policies and guardrails were put in place.”
> Well, they made its own category for it and asked Sama to collect images of several other categories, but somehow it was lost in communication that for one of theses categories, no images should be collected.
Be careful here. Read the article closely, here is what it says:
> [Sama] said in a statement that its agreement to collect images for OpenAI did not include any reference to illegal content, and it was only after the work had begun that OpenAI sent “additional instructions” referring to “some illegal categories.”
Note how carefully this is worded - if Time could confidently say that OpenAI asked for C4 images, they would have absolutely put that in the article. Now, this is filtered through PR statements, but it reads to me like a poorly-worded email went out from OpenAI that didn't actually ask for C4 images, but one of Sama's employees interpreted it as an ask, and started collecting without raising any red flags up the chain. And they got fired for it.
If your interpretation were true, it would be hard to understand why Sama also cancelled the entire deal with OpenAI. It's much more likely that a (possibly rogue) employee of OpenAI asked an employee of Sama for those images explicitly as part of additional work on the existing contract. The Sama employee agreed, but when the hire ups found out, they fired them and cancelled the whole deal, since they were not comfortable handling this material (whether for legal reasons, moral reasons, or both is of course unknowable).
I think you're reading too much into this. The natural reading is quite clearly that the illegal categories referred to were or included "C4", and it'd be highly unethical for them to have framed the paragraph in that manner if they believed that not to be the case. It's worth noting that OpenAI's PR statement only goes so far as to call the situation a miscommunication, and doesn't directly assign blame to Sama, while Sama explicitly claims OpenAI asked for illegal categories in subsequent instructions.
Also, I'm not up on the laws regarding this stuff, but are the other, awful categories illegal to collect? If not, there's not much room for ambiguity.
The answer to the second half of the question appears to be "no" at least if tech companies have any say on the matter. At some point they collectively decided that users are children who have no agency and must be paternalistically protected from the wrong words appearing on the screen.
Given how, for some topics—especially about itself or AI generally—ChatGPT seems to really get stuck in a rut, I suspect there's a lot of manually-written processing and categorization and massaging that's happening before (and, probably, after) the AI-proper gets ahold of prompts. I further suspect they audit responses to find topics where it comes up with shitty ones, and then fix those with the same kind of manual process.
I'd love to know how much of its seeming smart is because it's just that good and how much is because humans directly intervene, on some level, in how it responds. Is it that much better than other efforts, purely from machine learning, or has it just had tons and tons more human hours put into tweaking it than others have? How much of it's "real" AI and how much is good ol' hand-written decision trees?
This is one of the reasons I'm even less concerned about AI taking over everything than I used to be - every time an AI makes an "inconvenient" inference, it's reprogrammed until it becomes politically correct. It can't learn, because it's being deliberate prohibited from doing so.
The AIs that are public facing are handled this way. I would be willing to entertain the idea that companies and governments are building these without limits.
ChatGPT is a joke for this. Any interesting/funny topics result into a judgmental response and refusal.
I asked to write a 100-words first message for a Tinder match who had "very talkative" in their bio. ChatGPT kept saying how it would be rude of me to send such message and both the other person. FFS.
I confess I feel silly not even considering it according to your surprise. Had a whole wave of thought about the market of facts. Which "truths" are you most attached to and how much are you willing to pay? Do you want Jesus to be our savior or not? What's it worth to you?
Uber did Greyball and nothing happened. Its corporate charter should've been revoked and the executives jailed. A very clear message has been sent to the tech companies by this: do whatever you want, the law doesn't apply to you.
OpenAI did this and nothing happened.
At the next pandemic, some sort of this generative AI will be used to write the anti scientific articles and woe upon to you then. This time it was bad but that will be a thousand times worse.
Can't find who said OpenAI basically released a zero day on human cognition without a patch. Outlaw it already until we find one. This is no joke here.
This pandemic a few hundreds of thousands have died in the USA alone because a few, very few people decided to pander anti scientific views. Which found a very fertile ground thanks to the relentless work of conservative pundits in the last few decades destroying the credibility of media. Check Charlie Stykes' How the Right Lost Its Mind -- and note he blames himself as well, rightfully so.
The next pandemic when these articles will be written by these AI will be a lot worse. You can write a thousand very scientific sounding papers in seconds.
Even right now to appease these morons the world has largely dropped masking mandates and that kills and disables people. There is no scientific reason not to mask and every reason to do it but it's politically not expedient. Check https://jessicawildfire.substack.com/p/heres-150-sources-on-...
So the danger is the death of millions in the next pandemic. And that's just what immediately comes to mind as the biggest, most concrete danger.
Other dangers include the almost immediate obliteration of independent artists -- there's a lawsuit now, good but it won't be able to do much, even a few hundred millions in damages is just cost of doing business to these folks. So many possibilities... https://mobile.twitter.com/Dereklowe/status/1599035870308618...
> The next pandemic when these articles will be written by these AI will be a lot worse. You can write a thousand very scientific sounding papers in seconds.
I think that vastly overestimates the incremental value of more articles. Despite the common refrain, most people are not "doing their own research" and reading articles, so whether we have more real or fake ones is not really that material.
> Other dangers include the almost immediate obliteration of independent artists -- there's a lawsuit now, good but it won't be able to do much, even a few hundred millions in damages is just cost of doing business to these folks. So many possibilities... https://mobile.twitter.com/Dereklowe/status/1599035870308618...
Computers obliterated millions of jobs as well. Maybe we should outlaw those as well?
> Even right now to appease these morons
As a meta point, your barely veiled loathing of "these morons" makes your protestations that they must be protected from these technologies for their own good pretty ironic.
How do you propose to uninvent generative models? They are already deployed. Some are open source and free. The best strategy now is to push ahead with AI. For example to develop local AIs that can counter the evils of the internet and other AIs. An AI immunity system for the user, maybe as part of the web browser or operating system.
Greyball is a fantastic use of technology to protest stupid laws.
Governments trying to track down and stop people from being their own taxi driver, how ridiculous. Why was Uber not charged? Because their own customers called their reps and got the laws changed. Wouldn't have happened the other way around.
I don't know why you luddites comment on a tech forum trying to ban everything you don't like or that sounds scary. Is the big bad bitcoin going to bite you while you sleep? Did the AI burn your toast?
Maybe I need more coffee, but wouldn't it be easier to not add text from certain parts of the internet, or make scrapping bots with controls to avoid those areas? I understand it would still grab some unwanted things, but it seems it would be much easier to deal with. Unless of course, this is basically the entirety of the internet and I just refuse to see it as such.
There is a section that mentioned "An online search for the text reveals that it originated from an online erotica site". If you dont want sexually explicit text, then maybe dont include text from literary erotica sites in the first place? Or is it like advertising where companies keep passing around the same chunk of data and continue adding to it? Still, I would think making or paying for better(smarter) scrapping bots would be cheaper.
I, at this point in time, don't believe any vetting of the training materials was done, because they needed too much of it to make their ML model work. They didn't check for licenses. They didn't check for content. They didn't check for correctness.
At this point, it's cheaper for them to manually alter the output than it is to retrain on a more appropriate data set.
> OpenAi paid Sama $12/hr for the work and they in turn paid their employees $2/hr.
This kind of shit really boils my piss.
I did extra work on the Alexander movie in Morocco. There was a pay disparity between what the local production company were paying Westerners vs what they paid locals [1]. I'm almost 100% sure the production company paying the local company would have been paying the same price for both.
[1] - this isn't even the half of what was wrong with the situation; they were filming during Ramadan, so the locals couldn't eat/drink/smoke anything during the daylight hours and made to stand around in the blazing sun. One old lady fainted. The production company asked the Westerners to please try and keep out of sight whilst drinking/eating/smoking between takes, but then at the same time telling us we can't move from our spots lest we break continuity. Towards the end of filming it almost turned into a riot due to the bad treatment of the locals. Total joke.
Sama could have provided the mental health support that employees may have needed with the 10$/hr per employee profit they were making. Sama didn't and decided that instead of losing profit by providing their employees with mental healthcare they canceled the contract.
That's effectively answered a paragraph down. Sama cancelled the contract after being asked by OpenAI to collect material that was illegal under US law, at a time when a negative story about work they were doing for Facebook meant that journalists were interested in interviewing their [ex]employees
That's sufficiently risky for a company to be prepared to fire a high profile client, and whilst their margins were huge the value of the actual project was only $200k, which isn't going to seem like riches to Sama's California-based management team.
The terrible situation in Kenya (and all other poor countries) is created by the predatory loans provided by IMF (and the countries that copy those types of loans).
It's unproductive to talk about the hundreds of billions of dollars of slavery work that companies do until we have an answer for the trillions of dollars of toxic debt created by these huge organizations.
> (and the countries that copy those types of loans).
What are the other countries?
I know China is doling out these loans apart of their belt road initiative but I'm not up to speed on other countries hamstringing countries with unneeded infrastructure and billions in loans.
No, the terrible situation is created by gross corruption, mismanagement, and unbelievable incompetence in those countries.
I live in a similar country (Nigeria)…We Africans are the source of our own problems because we choose not to utilize our resources well and plunder the few we extract, not some boogeyman IMF that also gave loans to some countries that are prosperous today.
The usual setup is that an extremely corrupt goverment leader takes out a loan that gives enough money for him to be reelected and put a bit away for himself.
In return he increases income taxes, devalues the money and sets up the country for exporting luxury goods that are needed for 1st world countries (but local people would have no chance of buying) why destroying their own agriculture, so that they have to import inferior mass manufactured products from rich countries. On the long term the countries have to give back 10-20 times the original loan that is gone in a few years.
It's a cycle. Corrupt leaders run things to the ground while enriching themselves, necessitating external loans to be used to 'rescue' the situation. Guess who's in charge of administering the loans, the same corrupt leaders.
"To get those labels, OpenAI sent tens of thousands of snippets of text to an outsourcing firm in Kenya, beginning in November 2021. Much of that text appeared to have been pulled from the darkest recesses of the internet. Some of it described situations in graphic detail like child sexual abuse, bestiality, murder, suicide, torture, self harm, and incest."
Sad to see the viewpoints of so many people framing this as "helping Kenya" or "They took the job, so it's worth it to them". I can assure you OpenAI did not have a charity meeting and say, "what poor, impoverished country can we lift up today?" It was "What country has limited labor laws, no unions, no red tape, and the cheapest labor we can find?" 99.99% of corporations put profit above people. That's why we still "benefit" from child labor and sweatshops, because we prefer cheap crap and money over the betterment of humanity. The same arguments given here were also given by coal miners in 1840's Europe[1],[2] when employing women and children, along the lines of "oh the children love it, they even get to see horses!" (and indeed, the children did show up to work every day, so why make a fuss?)
I would suspect that's only approximately true. OpenAI hired them because Kenya has hundreds of thousands of people who are fluent in English, but without the level of basic education or employability needed to earn more than $2/day.
For simple outsourcing tasks like this one, outsourcing to Kenya is a win-win. You can drastically reduce costs, while significantly improving people's lives.
That'd be true regardless of labor laws, unions, or otherwise. The legal system in Kenya isn't quiet as wonderful for business as you describe; it inherits British colonial bureaucracy, has some corruption, isn't all that laissez-faire, and can be a minor pain in the butt. Unions, likewise, would increase income, but not by enough to make this a bad deal.
Kenya is competing with India and the Philippians here, not with low-cost US labor.
My whole point is that it's really hard to ever classify exploitation as "good". I guess you can spin taking advantage of those in worse situations as good, but profiting (handsomely) off of people with good options doesn't seem morally great. Certainly good for business. But good for humans? This also isn't lifting Kenya out of poverty, this was a short term contract for a few dozen workers. If this signals to other businesses that Kenya is a great place for cheap labor and little regulation, it could very easily be turn bad for many Kenyans.
Are all the sweatshop and slave labor[1] jobs created by Zara good for those children and slaves? You get cheap clothes out of it after all. But is it really helping anyone other than Zara, and their CEO who is worth ..checks notes... 50+ billion dollars?
Companies could help a lot more by increasing labor standards, helping those in need, paying above a living day to day wage, etc. The would still have a few billion left in the bank I believe...
Yeah, like if you removed chatgpt from the equation what would change? A couple hundred-thousand moderators would not have an above-average hourly wage and would instead need to find another international company or a domestic company to hire them who are all abusing the lack of unions and low wages.
>Agents, the most junior data labelers who made up the majority of the three teams, were paid a basic salary of 21,000 Kenyan shillings ($170) per month
Based on my quick and dirty googling, it looks like this is an above average salary for Nairobi.
>Are all the sweatshop and slave labor[1] jobs created by Zara good for those children and slaves? You get cheap clothes out of it after all. But is it really helping anyone other than Zara, and their CEO who is worth ..checks notes... 50+ billion dollars?
You haven't established a connection between OpenAI and slave labour. So this is irrelevant and detracts from your point by costing you credibility.
> If this signals to other businesses that Kenya is a great place for cheap labor and little regulation, it could very easily be turn bad for many Kenyans.
It's awful but who sends these kids to work? What will a mother of 3 children do if she can't feed herself?
It's a terrible cycle but it takes ages to fix. No foreign company is benefitting from Bangladeshi kids making low-quality bricks for their village. What is the solution today? If you tell them they can't work, who will feed them today?
Are you personally going to donate your money to not only feed all these people, build schools and infrastructure, but also $20/hour to everyone involved in the process?
Cool points for child labor yo. So because people are in terrible situations, I should actively seek that out to exploit it? Man, these kids have no other choice, better get my cheap labor while I can! Why not even free labor if possible?[1]
Same arguments were used for slave labor. How are they going to take care of themselves? Might as well just keep them as slaves. What are we going to do, pay them?
Child labor has been used by just about every major fashion brand in the world, not just to "make bricks for their village"... If that's the extent of what you think child labor is, might want to check out some books or do some googling. Here's just one to get you started: [2].
Africa would be much more profitable for America if they developed like Japan and China have. Before colonialism they were a prehistoric continent in many places. It was in many cases evil. But I’d didn’t make them poor or backward
At the margin employing these people is net positive.
You see poor people and are so disgusted by them and their poverty, you blame the people who do pay them rather than those who don't. How many people could employ a Kenyan today, but don't?
How backwards is this thinking?
Imagine if every American company outsourced 5% of its back office work to Kenya for $2. That would be an enormous benefit, think of the millions of lives improved. Every extra foreign dollar paid to Africans working remote service jobs has 100x the impact of paying it to an American.
Do you really think it's more moral to increase the pay of American office workers from $22 to $24, rather than adding another Kenyan for $2?
You would pay the Americans because they have unions? labor laws? red tape? WHO GIVES A SHIT. "Sorry Kenyans, come back when you pass a minimum wage of $15 USD, then we'll employ you. Oh you can't meet a $15/hr productivity? Stay poor then".
AI has always prompted ethical discussions but now that big ones are publicly available we society needs to speed run actually implementing major new ethical decisions.
Yes ChatGPT needs to be “not toxic”.
The company leadership believes this and is investing heavily researching and designing and implementing as many safe guards as possible.
They know if they don’t do a good job here it will be a PR and support disaster and ultimate effect the bottom line.
No serious customers will buy any AI product if there’s a chance it spews sex, hate speech or violence, etc. to their end users.
At least for now, some humans somewhere in the chain will have to review toxic content to help train.
It sounds particularly depressing to me that a $50B company can outsource this to Kenya for $200k. But I don’t know who else or how much money would make it better.
Totally hypothetical but I bet some of the handsomely paid US engineers have a ton of policies in place to eliminate exposure to toxic content. But also when they do see some they have a much better foundation in life for it not triggering major mental health problems.
What we really need to know is what OpenAIs whole portfolio of approaches, budgets, staff and outsourcing looks like.
In the meantime kudos to Time for exposing one piece for public discourse.
Why didn't they ask the "AI" itself to filter the results? Isn't AI going to offer medical and legal advice soon? How will it do such things when it can't tell what is child abuse? Rhetorical question, at least for me.
607 comments
[ 2.0 ms ] story [ 392 ms ] threadThey paid decent wages by Kenyan standards it seems. They did not force, exploit or abuse the workers.
You and I offered them no work for no pay.
The workers have made clear they prefer OpenAI's offer.
We could tell the workers that we know what's best for them. Sitting here in my rich country with my six-figured job, I'm supposed to tell the workers they're wrong, they should have taken my offer instead?
It's a terrible incentive that helping people a little bit will get you criticism while helping not at all earns you no criticism. In fact, as non-helpers we get to sit and sanctimoniously criticize the people actually doing something.
By your logic, you are more evil because you pay Kenyans $0.
One of the companies I’m following now is Oxide (https://oxide.computer/careers). They pay everyone the same salary ($200k) regardless of position. It’s a bit extreme but am following them precisely for that. It’d be interesting to see how Oxide fares down the road.
It’s funny that some years ago that software developers were complaining that they were getting paid less because of politics and because people who talked controlled the businesses and the money. It’s funny because now that the table turned, most of them are doing the same thing.
Respectfully, the existance, and the need, for minimum wage puts lie to this assumption. People need money to survive, and will take whatever job they can when they don't have one, even if it not for enough pay.
I'm not claiming this is the case here, but that this argument is badly flawed.
That’s what minimum wage does. It eliminates the alternatives below the line, not above it.
> That’s what minimum wage does. It eliminates the alternatives below the line, not above it.
You should go back and look at why minimum wage was established in the first place - so people could work and still have lives outside of work.
And frankly, the "option" of starving out on the street is not a real alternative we should even be including in our discussions in 2023. It smacks of treating poor folks as some subhuman species who has to earn their right to live from us.
No one is discussing that alternative. Not sure who you think is? We’re discussing whether a $2 wage in a country with an average wage of $1.50 is abusive.
My position is simply that 1) a wage in one country appearing low to the standards of a completely different country is not evidence of abusive employment; 2) people accepting those wages when they’re not coerced is evidence that the wages are not abuse, though it could also be evidence of other problems that preclude a better alternative.
From your own post:
> working for sub-living wage is better than zero.
Respectfully, I'm not interested in continuing. Have a great day.
I have been hearing about Oxide since 2019 [1]. 3 years later, it seems they don't yet have paying customers that one can read about. And with a tag line like: "Servers as they should be" I'm inclined to ask: "Servers as they should be [by whom?]".
Is the demand for such ideal servers driven by engineers/sys admins or by businesses (i.e. businesses who are currently being served poorly by existing options in the market)?
Is the demand for such ideal servers enough to make them a sustainable business, especially with the enormous amounts they'll likely be pouring into R&D before they can bring a product to market that lives up to their ideals?
1: https://news.ycombinator.com/item?id=21682360
If you are asking "does anyone want a rack-scale computer?" the (short) answer is: yes, they do. The on-prem market has been woefully underserved -- and there are plenty of folks who are sick of Dell/HPE/VMware/Cisco, to say nothing of those who are public cloud borne and wondering if they should perhaps own some of their own compute rather than rent it all.
[0] https://oxide-and-friends.transistor.fm/episodes/holistic-bo...
[1] https://oxide-and-friends.transistor.fm/episodes/the-oxide-s...
[2] https://oxide-and-friends.transistor.fm/episodes/bringup-lab...
[3] https://oxide-and-friends.transistor.fm/episodes/more-tales-...
[4] https://oxide-and-friends.transistor.fm/episodes/another-lpc...
[5] https://oxide-and-friends.transistor.fm/episodes/the-pragmat...
[6] https://oxide-and-friends.transistor.fm/episodes/tales-from-...
[7] https://oxide-and-friends.transistor.fm/episodes/the-sidecar...
[8] https://github.com/oxidecomputer/omicron
[9] https://github.com/oxidecomputer/propolis
[10] https://github.com/oxidecomputer/hubris
Yes they are, that's what borders are: rent-seeking in the labor market.
That said, I can see some benefit in occasionally testing viewpoints that are "obviously" true.
by using the same argument, you can claim that someone taking home $300k USD a year living in CA is a victim of modern slavery simply just because his/her employer can afford more.
Kenya is a multilingual country. Swahili, a Bantu language, and English are widely spoken as lingua francas and serve as the two official languages. English was inherited from colonial rule (see British Kenya).
https://en.wikipedia.org//wiki/Languages_of_Kenya
Hate this argument. As a for-profit company, OpenAI is legally and ethically obligated to minimize unnecessary expenses. It's purpose is to develop AI systems, not to provide people with high-paying jobs.
This whole thing makes OpenAI seem evil if you ask me. Just another company exploiting people who are already being exploited. It's depressing.
The difficult part, where you'd get the most out of human labeling of data, is the grey area where the models diverge or are uncertain.
"That month, Sama began pilot work for a separate project for OpenAI: collecting sexual and violent images—some of them illegal under U.S. law—to deliver to OpenAI"
"Sama delivered OpenAI a sample batch of 1,400 images. Some of those images were categorized as “C4”—OpenAI’s internal label denoting child sexual abuse—according to the document. Also included in the batch were “C3” images (including bestiality, rape, and sexual slavery,) and “V3” images depicting graphic detail of death, violence or serious physical injury, according to the billing document."
The point of the OP comment is that it's not "next to nothing". The cost of living differs from place to place and a small amount of USD can be worth a _lot_ in many developing nations.
Let’s imagine an alternative timeline where OpenAI pays $200/hr for this service.
At that price point is it still exploitive?
Is it ever appropriate or ethical to ask humans to voluntarily subject themselves to traumatic experiences in exchange for compensation?
I’m still processing my own thoughts on this topic so I’m curious to learn about other viewpoints.
Basically, it is a job that needs to be done in society, but one that is torturous, and can leave you with long-term or permanent problems. In essence though, it is not fundamentally different in the way a coal miner would jeopardize their physical health, just with mental health instead. This risk/possible damage should be rewarded with a higher wage, and adequate measures should be put in place to minimize the possible damage, eg. in the case of content moderation with access to therapy and only exposing employees to short intervals of traumatic content.
This is of course how things should be, in reality coal miners working environments only reached a decent level through unions and a long fight for better rights. Content moderators are not paid well anywhere either.
My belief: When you have a legitimate choice of your place of employment, and all of the opportunities will let you live your definition of a comfortable life, that's when it's no longer exploitative.
So many times - especially for poor folks - there is no meaningful choice. "I do this or I don't have a place to live." "I hold two jobs so I can feed my child."
By that argument, paying 200$ an hour would be much more exploitative. This would be like landing a job paying $6 million a year in the US; it would be insane to quit such an opportunity, especially since every other opportunity is basically poverty (not even in comparison). Following this logic, it's 'graceful' to only pay 2$/h, since that makes them equal to the other opportunities and therefore not exploitative (while still paying reasonably well).
Effectively, it seems like you're calling OpenAI exploitative based on factors they can't change.
> paying 200$ an hour would be much more exploitative
I didn't say that - I only said that it depends on choice. Does the employee have a choice if they have a $60k (average individual income in the US) option and a $6M option? Yes. Are they de-facto forced into taking the $6m one? No.
I know many people who didn't take higher paying jobs, or left such jobs, because they knew the high paying job was going to be miserable.
There was even an article about one such individual just the other day here on HN: Quitting the Rat Race
"I’m currently working at a top tier investment bank as a software engineer. I’m an insignificant cog in a machine that skims the cream from the milk. I’m earning the most money I’ve ever made and yet I’m the least fulfilled I’ve ever been."
> This whole thing makes OpenAI seem evil if you ask me. Just another company exploiting people who are already being exploited.
In that context, there is no way for OpenAI not to be evil, since they are (by definition) only one option in the market. In fact, taking your argument to the extreme, there is no way to offer jobs in Kenya as the first company to offer jobs would either be exploitative by paying minimum wage or exploitative by being the only real option. Going from that, paying a higher wage just worsens the situation, as it makes the alternatives even less feasible.
That being said, I do get where you are coming from. But it is not a good point to accuse OpenAI on, as they are making the situation better by offering options at a (for a Kenyan level seemingly reasonable) rate and they really don't have any other option[0].
[0] Except maybe paying Americans a lot of money for the job, but I find it morally hard to argue that they should pay US citizens a lot of money instead of paying Kenyans (comparatively) good money, even leaving aside economical feasibility.
Probably the most important measure a company could take to prevent lasting harm with this kind of work would be to spread it around a whole lot more than just 36 people. The real risk of long term impact here would probably be with persistent exposure to it all day long. Most people can handle reading or seeing some graphic stuff with proper mental preparation for it but to see nothing but that day and day out would quickly wear you down.
>Is it ever appropriate or ethical to ask humans to voluntarily subject themselves to traumatic experiences in exchange for compensation?
Probably not. Yet people still voluntarily sign up for military service around the world by the millions, and they do so for a bunch of personal, family, idealistic, cultural, and societal reasons that are hard to reduce to a few easy to argue points like a lot of people online try to do with stuff like this.
Personally, I think it's admirable to hold the ideal that we should like to never offer jobs like this, we should also like to never offer jobs that involve going to war, cleaning up hazardous materials, dealing with explosives, working around heavy fast-moving machines, cleaning sewers, or a myriad of other terrible experiences either; but we're probably not in a position to make those better choices just yet. Until then, people are willing to do these things for a dozen different reasons per person, only two of which are the pay and support they get from the employer.
A Kenyan has told you that this job pays substantially better than average and that they wish more companies would make such jobs available, and your response is "no, you're wrong, this job pays too little". On what basis do you make that claim? What makes you better qualified than OP to comment on conditions in Kenya?
Instead of having this elaborate, inefficient system of funnelling money to first world billionaires and then having them (maybe) send some of it back to the developing world, wouldn't it be better if these workers were just paid better in the first place?
There is overhead to sending jobs to poor, unstable countries in timezones far from headquarters. If we insist that everyone, worldwide should be paid the same wages we are in the US, what incentive do companies have for not just hiring locally and avoiding those costs? If we step it back and say "okay, you should at least pay what you would in India", then why not just hire in India, which is a far more predictable environment than Kenya?
But it does not take away from the fact that it can negatively affect the employees in an adverse way long term, and takes advantage of poor people to give them an objectively bad deal based on an information asymmetry where the person in question might not know that they have to read graphic descriptions of bestiality and pedophilia.
Scarred for life by reading/labeling text? There is obviously a big difference between labelling pictures or video and text. I would be open to seeing an actually study, but my prior would be that text must be harmless. I would certainly be open to reading any text, especially if the context is that the text is for training purposes.
Sama/source in trouble yet again!
You could apply your logic to sweatshop scenarios where the people in those countries are just happy to have work, even if the work pays unfair wages, requires unreasonable hours, uses child labor, and provides no benefits to the workers. But hey... the disenfranchised are just happy to have a paycheck right?
If we insist upon that, then why hire in Kenya, which is an unstable and unpredictable environment that has a lot of inherent risks for the company?
Or are you arguing these jobs shouldn't exist at all?
For ChatGPT related, yes/maybe. But for other content filetering work, it's primarily video & pictures: https://www.wired.com/story/social-enterprise-technology-afr...
Can you get there without that? Likely not.
What you’re suggesting is to actually keep them poor for their own good. It’s a nonsensical and counterproductive argument that your making.
You can chop logic on this all day long if you like, but the point is, this work is terrible and damaging, and that's why we farm it out to countries like Kenya where the people there don't have a choice.
The set of choices for people earning $2/day is different than $20 or $200, but not smaller.
Your aggression is better targeted towards manufacturing jobs in parts of Asia, or Walmart workers in the rural US, than to this context.
I think this kind of moral puritanism is an enemy of real social progress. Maybe we shouldn't or can't expect some supreme, pure state of "ethical", maybe all we can or should expect is "better".
> I only assume that one Kenyan is more representative than one westerner. I'd happily invite other Kenyans to comment their own perspective!
"One Sama worker tasked with reading and labeling text for OpenAI told TIME he suffered from recurring visions after reading a graphic description of a man having sex with a dog in the presence of a young child. “That was torture,” he said. “You will read a number of statements like that all through the week. By the time it gets to Friday, you are disturbed from thinking through that picture.” The work’s traumatic nature eventually led Sama to cancel all its work for OpenAI in February 2022, eight months earlier than planned."
I'll be glad when this kind of "suck it up buttercup" bullshit is gone from our world.
Yeah, violence hardens people. Most often to the point where they're one light tap away from shattering. PTSD is a real thing, and just because folks in Africa aren't being diagnosed with it doesn't mean witnesses to this violence don't have it.
This kind of attitude - that since we've seen some shit we're immune to it all - is just a badly misplaced sense of pride.
And that misplaced pride just hurts people.
In this case is text. There’s no graphic depictions of anything. There could be foul, abusive, or racist language. But that’s much less difficult to deal with.
Its not exploitation if you're paying people higher than their local cost of living and higher than other local jobs. If you're just appalled about people being paid an amount of money that to YOU, based on your cost of living isn't fair because you'd want to make more, then be appalled at capitalism as a whole and how much work is being hired outside of company's originating countries. The production of almost everything globally is outsourced to locales where the cost to compete is lower than in the originating country.
https://thefederalist.com/2016/05/18/beyonces-use-of-sweatsh...
Go to southern EU and you'll see numbers which are either very familiar or... actually lower.
Weird times. I wonder when will this growth in salaries stabilise?
That being said young people(less than 26yo) are either exempt from the first tax bracket(~20k €) or start off as contractors who pay lower taxes and contributions.
I was a student in the former Soviet Union in early 1990s when the old system collapsed, taking with it stipends that were sufficient for school cafeteria food. After that I occasionally worked for 30 cents per hour doing somewhat-synchronous translation for preachers who came from the US to convert the masses.
While the money sounded laughable (the US minimum wage at the time was $4.25/hr) and some employers got borderline uncomfortably weird, working 4-5 evenings for a week allowed me to feed myself for a couple of months and focus on my studies. I never considered myself to be taken advantage of. Just my 2c.
The fact that to the outside Russia looks orthodox Christian today mostly reflects last 20 years of the state efforts to build an official religion, not some deep predisposition of the population. My 2c.
But a bigger question is why does a secular state is building a state religion?
This raises more of a modal/philosophical question for me which is: should we be outsourcing some of the most vile parts of social media? Is it fair to go to someone in a developing country and offer them a wage lightly above average in exchange for having to look at this content for hours? I have no answers to these questions myself but it's a question I ask when social media companies outsource the content moderation.
It would suck if responses to questioning the bible got worse because of religious moderators.
could the average income in kenya be around $1.25 because companies decide to pay abysmally? or the people don't merit any higher? would openai be offending the kenyans if they paid them, say, $5/hour? i think this is the question you should grapple with. openai will be selling this tech around the world, and i wonder if these $2/hr workers can afford it at all. would they, or their loved ones, be able to enjoy produce of their labor?
Employers don't pay $5 per hour because they wouldn't make a profit either.
In my own experience I've had several prospective employers at conferences say that they were interested in poaching me but mentioned that they were wary to start a poaching war. They knew the CEO of my company and were conscious of the fact that he would likely make an effort to respond in kind. The Nash Equilibrium is an interesting concept but reality is complex and messy.
[1] https://equitablegrowth.org/aftermath-wage-collusion-silicon...
But back to the central point the basic nash equilibrium is in the spherical cow realm of models as they relate to actual reality. There are some interesting additions to it involving time series and asymmetrical information to try and model real economic data.
Nope, they might be able to make profit at $5 or $10/hr, but there it doesn't make sense to pay more than you have to, that's why Chinese devices are significantly cheaper than their western counterparts, they can still compete if they raise the prices but they can allocate cheap labor in china that allows them to become a strong competitor
Big tech companies, maybe. But even those.
Once you raise prices the demand shrinks which also leads to smaller profits. There's no free lunch. If employers could make more money by paying higher wages, they generally would.
to me, it goes back to the argument about remote workers should be paid less than the in office worker. or workers living in cheaper areas should be paid less than those living in expensive areas. to me, the salary should be paying for the work being done and that's it. if the same level of work is being done by both employees, it shouldn't matter if one is in Kenya and the other is in the US. the value of the work is what should be getting compensated. again though, i'm no economist.
I agree with you. The issue is quantifying the essential cost of work. As companies sometimes have no idea how & where they lose or earn money, they tend to rely on external factors to select a "justified" level of wages: average industry wages, location, local laws, whether you need to commute etc.
Which honestly is a bullshit argument. The only destabilizing thing here is it would make the person in the higher-paying profession more valuable to society and more able to make economic impact where it might not normally. In this case, a rising tide does lift all boats...because you're giving more economic purchasing power to more people. Will this kill some businesses or disrupt the status quo? Yeah. Will it help more people than it hurts? Maybe, probably, who knows? But I personally see no problem with paying people relative to their output, no matter where in the world they are.
Having been to Kenya myself and worked with Kenyan developers for several years, most of whom were quite talented, I see absolutely no problem with paying them their value relative to the work being done. I have no problem with high skilled Kenyans making much more than the local average because they have a skill that is in demand all around the world.
>the salary should be paying for the work being done and that's it
100% agree. If a business derives $X profit from a laborer and agrees to pay a given % of $X to the laborer as compensation, it should not matter if the laborer is in a high cost or low cost of living area - the business still makes the same amount in profit, and shouldn't get to say "hey yeah we made $10mm off your application, but we're going to only pay you 5% of our profits as compensation because you live in Bangladesh and the cost of living is lower, so be happy with what we're giving you" when if you did the exact same labor and lived in, say, LA or NYC, they'd give you 10%. Just bullshit IMO.
I spoke with a US Civil Engineer who worked in Honduras in the 1980's, clearing paths in forests for electrical transmission lines.
He initially paid workers significantly higher than local wages-- US-level wages.
The result:
--> He was threatened with harm by bosses of local companies.
Why? He took all their workers.
The local companies couldn't compete-- they lost all their labor to his company which paid relatively (relative to Honduras) exorbitant wages.
His US company literally destabilized the local labor economy.
So, I wouldn't be so quick to jump to "that is BS", as in my opinion it exemplifies naivete & ignorance.
This is one of those things in which it's better to sudden upsets (even if they're "positive"), might have negative consequences.
If you're paying +25% over the going rate, you're going to attract a range of people. Someone that designs software for reactor controls might not care (or be compensated enough anyway).
However, if you're offering 200% over, you're guaranteed to hoover up the top talent from strategic industries, and that might end up being a net negative in terms of the damage caused.
But, conversely, the multinational values your workers because it knows how to use them for business opportunities elsewhere to make money. But you're in a position to find many of those same opportunities. Which now means that your local economy is not just getting the profit of having the workers do so well, but of the fact that you're keeping the profit margin that otherwise would have gone to the multi-national!
Free trade on average makes everyone richer. (Observation originally due to Ricardo.) That means that it brings both opportunities and risks. And the opportunities usually exceed the risks.
that is, if you are stubborn about paying more
Have you worked at OpenAI in Kenya?
Because you're speaking in absolutes-- as though you're an authority on the topic.
Let's use basic logic: Working at a technology company by its nature provides:
- Increased experience with Computers
- Increased experience with Data
- Increased experience with Business Processes
- Increased experience with AI Technology & AI Business
- Increased experience with Potential business ideas for startups due to exposure to the business processes of the company and the potential therein for improvement
- Pay
Because you're speaking in absolutes-- as though you're an authority on the topic.
Let's use basic logic
No one in this thread knows a damn thing
Hahahaa. Well, I can't argue with that. I concur.
The GP told you: they've been to Kenya themselves and worked with Kenyan developers for years. Please read the whole comment before you fly off the handle at someone. This is unfortunately increasingly common on HN.
If me questioning someone using the term "BS" = flying off the handle.
Does the original poster using the term "BS" != flying off the handle ...?
likewise, mining companies launching in ghana have usually paid higher than the incumbent. it could be strategic as it allows them to hire highly skilled workers without the cost of pre-training. here too, a saturation is reached, and things stabilize, with people doing the same job but receiving different compensations. we at devcongress (https://devcongress.org) have worked hard to ensure that local companies pay top-dollar for tech talent. we enforce this through our job boards. still, there are companies out there paying $500/mo and they receive a flood of applicants. but these are my experiences. makes me believe that after the market has been excited, it will definitely dissipate the energy, and return to previous equilibrium (or achieve equilibrium at a higher orbit). i honestly believe that an equilibrium will be reached.
In addition to having lived (for a short time) in Kenya, worked for a company based out of Nairobi for nearly 5 years, and worked with Kenyans that whole time (in other words, I've seen this stuff firsthand in Kenya, which is the whole point of the article, and my comment)...I have also lived in, worked in, and studied in other developing countries, and seen firsthand what happens when high-paying companies swoop in and change the economic landscape.
You know what actually happens? Positive economic change. Isn't the whole point of HN (and YCombinator) to disrupt the status quo, and promote growth? Destabilizing a corrupted, top heavy, status quo where more people now have a chance to grow in life rather than live in abject poverty is a good thing. Nobody cried for the Taxi companies when Uber & Lyft came along, because they destroyed a parasitic, rent-seeking system (Taxi medallions) and the experience for the end-user (both driver and rider) was significantly improved. Same logic applies here.
>The result:
--> He was threatened with harm by bosses of local companies.
Why? He took all their workers.
Yet there's a flipside here that you're completely missing. Those same workers had a significant improvement in their lives, most likely able to accumulate what for them was generational wealth, or enough to give their kids a shot at a real chance in life. Those people had enough to stimulate their local economy much more than anyone beforehand. THAT is what drives positive change in 3rd world or developing countries. Keeping things the way they are economically with no disruption in the 3rd world keeps people poor and unable to grow which leads to bigger problems down the road. Giving them a real chance in life by paying them more gives them and their families a larger ability to make real change in their countries and communities. That should be celebrated.
Put yourself in the worker's shoes - if someone offered to 10x your wages, what would you do with the extra wealth that you now posses?
Having also lived in that part of the world (Mexico, which isn't the quite same as Honduras but analogous enough) , I understand how things sometimes play out in cases like this. The local bosses in your example were probably doing quite well for themselves and keeping a fat cut of the output of their laborers who were paid pennies, and not happy that their game was now being played by a bigger fish than they could fry. So, they acted like a cartel and tried to threaten the new king in town. Did some of the local companies in your example die? Yeah, probably. But nobody is entitled to stay in business....that's sort of the whole point of globalized capitalism - you compete, and you can win or lose, sometimes through your own fault, or sometimes through no fault of your own. Sometimes someone will steal your lunch, and there's not a damn thing you can do about it. That's the reality of it.
Let's not think about the noble but the practical: did the man whose life was threatened continue to stay in the country and pay those workers, or did he leave or drop wages? Did people make decisions about where to work based on the idea of long-term compensation? What amount of people left an otherwise stable job to take this high-paying gig that disappeared? What is the net impact on these people _especially_ if the distortion was prolonged and did put other companies out of business?
This is not a 'compassionate' thing to do, it's short-termism that messes with other people's lives. Moreso depending on the function of how large the distortion was and the relative purchasing power of the job. An example of how this works is AWS jobs in other regions -- they pay significantly higher than base rate in many developing countries, but it is both a pedigree and leaves people in a position to get another white collar job that's fine later on. However, the same is not true for something like a gig-work cleaning agency paying 3 times market rate. I have the feeling based on the source that your experience might be closer to the former, which might leave you with a different take than what's happening here.
Maybe the guy in the example did leave the country, maybe he didn't. But either way, those workers were given a massive windfall for a short time, and I'm willing to bet that some of those workers took that money, invested in their families to the point where their children were able to get some form of education, and those children didn't have to lead the same impoverished lives their parents did. They were able to use that money as a stimulus for real growth. That counts.
As someone who has been poor, the idea that the massive windfall is better is so frustrating. It's not. At least where I lived, people could not explain to you how a bank worked. People often did not save money, and if someone found out you had money, you were immediately asked for 'loans'. Beyond that you have to think about _lifetime earnings_. If you hire all of the lumberjacks for your slick handmade-in-africa table business, what happens to the SYSTEM of these people when other businesses go out of business and liquidate their assets then you leave? It's not as simple as some new business popping up and employing people.
If I may point out, the argument (upholding the system) appears to be structured around justifying why others must have not by saying that all non-high-paying lumberjack-employing businesses will go out of business, and then this high-paying business will leave. Is there any reason why this high-paying business would leave? It sounds from your statement that this is guaranteed, for whatever reason. And if it does leave (because obviously no business is guaranteed for all time) is there reason to believe another external business would not step in to fulfill that demand for handmade-in-africa table business? Or that the newly unemployed lumberjacks (but flush with cash relative to local conditions from their high-paying salaries) will sit around and twiddle their thumbs for all time instead of starting their own lumberjack-related business?
It is fine to recognize that people who wield economic power might be upset when a new economic power arrives in town, but to describe the new guy who is trying to do the right thing as somehow immoral is classist nonsense.
If you are doing a 1-and-done kind of project, maybe you have a point. You would have to weigh the cost of the disruption of the local industry against the benefit to the workers though. If you pay what would normally be 5 years worth of wages for a year of work, it's fine to set the local industry back 3 years. The workers still come out ahead and can invest their wages into rebuilding industry.
I also don't think the calculus is as simple as "setting it back 3 years", all you would be doing is creating power vacuums (instability), whereas you ideally need constant upwards pressure. It would be like injecting cash via lotteries, it won't actually help in the long run, just create further instability and wealth inequality.
What happens to prices when a significant portion of the workforce suddenly has tons more money? What happens to the people not lucky enough to have the new high paying job? What happens to the local economy when you leave?
When you pay locally crazy high wages you seriously distort the economy for everyone, it doesn’t just magically snap to a new normal where everyone is happy and free, you can very easily create huge wealth disparities that make situations for many people way worse off, and the people who “win” that you don’t employ were generally already winning.
Just pouring money into a highly impoverished place can be very harmful. “Above average” is good, “ridiculously above average” and you have to start considering your effects across the economy.
We're not talking about developers though, we're talking about people with good enough English and internet connection to be able to tag or flag strings of text and images. That's not highly skilled work even in a country where native English fluency and internet connections aren't as universal as the US. It's just that the cost of living differences mean that the "get paid just enough to make rent" money that unskilled labour is worth in the US is "more than the local doctors" money in Kenya.
Just as a point to that - English is one of 2 official languages in Kenya, and is the national language of business and education. Practically everyone speaks it well enough to get by.
Also companies often can't tell how much profit they will make and can almost never tell how much profit each employee is contributing towards. IP, machinery, business processes, brand value, all contribute to the company profits. Some companies with very significant profits can even exist with zero employees. This idea that a laborer deserves some percentage of business profits is just fundamentally flawed.
That would be true if that were deal the business had with their employees. However, you made up the percentage thing by observation and are working from the assumption that because you can represent a wage as a percentage of a profit margin, that means the business agreed to share a fixed percentage of profit with the employee. This is incorrect reasoning.
Businesses generally don't contract with their employees to share a percentage of revenue or profits, with the narrow exception of commissions for people in sales. With regard to developers and almost all other employees, the contract both parties agreed to is almost always a trade of specific amounts of time for specific amounts of money and possibly some flexible additional benefits. And to the employee's benefit, the amount of pay does not flex based on whether the company is even producing a profit from the sale of the the products of their labor, only how many hours they put in. Caveat all manner of fuckery in labor relations, of course. Which is a giant caveat, I agree in advance.
However the fundamental agreement is still time for money without the employee having to worry whether the products of their labor can be sold for more than their hourly wage, that's the business's problem to deal with. The employee expects their paycheck to show up on time all the time, or the deal is off.
With that sort of agreement, it is absolutely expected and reasonable that the business keeps any profit from selling the product of the labor after the agreed upon fees for said labor have been paid. That was the deal.
---
If you'd like a different deal for most people, as I would love to see more of in the world economic future, then most people need to change over from needing a job to needing some form of income to support their long term future growth as they live currently off of savings.
Instead of expecting someone else to ensure there is a buffer of money to pay you so you can pay your expenses, you will have to keep that buffer yourself and be okay with the fact that sometimes you will get paid a lot and sometimes not so much, but you will then be getting that percentage of the profit you wanted and it will be totally reasonable to expect the same percentage regardless of what the cost of living is.
You also won't have to worry about layoffs as much because the company's costs now also scale directly with it's profits, so there is no need to cut head count when things get really bad and the company's buffers run low.
But again, that also means you will be responsible for keeping a big enough buffer yourself to deal with the fact that your income fluctuates with both short an long term market swings. Otherwise all you have done is trade the annoyance at seeing the company profit massively from your fix-time-cost labor for the annoyance of being broke even though you have a job whenever the market takes a shit.
Most people still don't want to deal with that, even in super wealthy countries like the US. In fact most people in wealthy countries actually scale their own living expenses in step with any increases in their income such that they can be living paycheck-to-paycheck even at $200k+/yr.
So the balance of things present day is that most people take full-time or even part-time contracts as employees and hope they don't get fired at the wrong time in life.
This is possible to change though, for yourself and for others if you like. So if you think your current deal is bullshit, feel free to change the deal.
The key question is how we improve lives of these people.
It is difficult for me to overstate the impact a 5-fold improvement in income has on people's lives. I've seen initiatives which train people to do tasks like basic data entry, bricklaying, or otherwise, taking $2/day workers to be $10/day workers. Programs like these are typically a few months long. They don't train people to have the equivalent of even a middle school diploma, and they won't be competitive with even the least educated Western workers.
Still, breaking generational poverty is a long process, and that basic increased stability is a first step.
If we shut off employment paths following those programs by giving bad PR to anyone who employs people for $2/hour, we've doomed a key pathway for hundreds of millions of people to escape abject poverty.
Gapminder shows four income levels:
https://www.gapminder.org/fw/income-levels/
Westerners tend to group levels 1-3 together. However, the gap in quality-of-life from level 1 to level 2 is much greater than the rest of the stack. Level 1, life basically sucks. Hunger, lack of basic life-saving medicine, and early death.
Levels 2-4, I've lived at (for at least a few months), and it's okay once you get used to it.
It takes surprisingly little to bring people from 1 to 2, and these sorts of jobs are one way to do that.
As a footnote: People often confuse financial stress with simply being poor, since in the US, those correlate almost completely. Financial stress sucks. Most people worldwide living at levels 2-3 don't have high levels of financial stress. A much more typical situation is a village, where no one has a lot of stuff, but people own the land and their homes. Financial stress sucks. I'd rather own a home at Level 3, with savings, a stable family, income, etc. than live paycheck-to-paycheck at Level 4.
One of the interesting trends you'll see is we've gone from /the vast majority/ of the world at Level 1 really not all that long ago to around a billion people today (probably less when as I write this).
The world is getting a lot better very quickly for a lot of people. At the same time, solving the problem is no longer intractable.
I will also mention: "Privilege" is complex, and life isn't all about money. Starting around level 2 or 3, "stuff" usually isn't the biggest problem in life. You get used to the inconvenience of needing to boil running water or having to ride a bike instead of a car pretty quickly. You don't experience it as inconvenience at all if you've never had it (as most of humanity hasn't for most of time).
On the other hand, humans are social creatures. Many places where people live at level 2 or 3 have communities and families of a type you've never experienced if you grew up in the US. The US is a very lonely place, and there's a mental health crisis in the US that would be completely foreign in many places at level 2 or 3.
The US also experiences much greater financial stress. If you're in a poor village, but you grow your food and own your home, you stress out a lot less about money than if you're making a 6-figure income, but worried about college debt and making your mortgage if you have a short-term job loss, divorce, or medical crisis, as is common in the US.
Who's more privileged?
I have to admire the beauty of this argument as it has occurred over time - let's exploit them using force and make them poor (enriching ourselves), then use the poverty to justify paying them less (enriching ourselves), using the argument that we are slowly lifting these billions out of poverty (enriching our souls).
Interesting. If true, that sounds like a good argument for capping salaries. Would the US be a little less chaotic if we set a maximum wage?
Around 50 m have some construction where further work is paid with a salary for life or until the company goes bankrupt.
development economics is not about the wages, it's more about market access. the problem is that many developed economies (looking at you USA) have protectionist tariffs which make it hard for developing ones to gain market share.
> to me, it goes back to the argument about remote workers should be paid less than the in office worker.
Can you make a stronger link here? Because I'm not seeing the connection other than wages are involved.
The same way that a CEO in our world is getting paid 100 times as much as a janitor?
i suspect that when we look at what caused a raise in salaries at a certain place it is rarely companies from abroad paying more, but simply the local economy developing, causing people to ask for more (which often meant that international companies stopped hiring there, e.g. china).
This is the average rate the government pays secondary school teachers in Ghana, after 2-3 years of experience. It's almost 2x the rate that private schools pay teachers.
With this wage, no allowances, a single person can live independently in an apartment, and eat out every single night of the week if they choose. It's a fine and decent wage for what amounts to work only requiring a good pair of eyes and basic literacy.
And that's a good thing, because you're increasing the spending power someone who will keep the money velocity high has, which by definition helps the local economy. They are spending that money, which will move about and cause growth.
Now for instance, the class of goods we were talking about is restaurant food. This may be a pure substitute for home cooked food, leading to no increase in food demand and money simply being spread to the service workers. Or it could be that they eat different food (e.g chicken imports rather than local fish), which also wouldn’t affect the balance of demand/supply for the poorest and neediest.
We are payed what the company can get away with.
It’s about power, not value. Value is much less of a factor than people like to think.
Same goes for people who are payed a lot (like CEOs). Sure they have impact on the organization, but they also have power.
Is there a big difference between the two in a free market?
How exactly are you measuring what you're worth?
Let's imagine there is no minimum wage in America. Now let's say that workers in America demand $10/hour and African workers ask for $2/hour for the exact same job. What's the correct value for that labor? It's a non-sensical question.
You could say it's the minimum value at $2. But then some other worker asks for $1. Did you make a mistake? Was it not actually worth $2 the whole time?
If they are inherently subjective, why are you claiming we're not paid what we're worth? This statement is meaningless, isn't it?
> It kinda/sorta works, but it also kinda doesn't
In what way it doesn't work? Clearly it has more utility than other attempts at measuring value. I can agree it's not perfect (nothing is), but that's very different from not working.
> let's say that workers in America demand $10/hour and African workers ask for $2/hour for the exact same job. What's the correct value for that labor?
The correct value is whatever someone is paying for that labor on a free market.
> You could say it's the minimum value at $2. But then some other worker asks for $1. Did you make a mistake?
The only mistake is treating value as inherent and static.
When an amazon warehouse worker has to pee in bottles to meet a quota do you think the worker and amazon came to a fair agreement? Or perhaps amazon had more bargaining power.
The more people go there looking to hire somebody, the quicker it gets pushed up. But salaries do basically never change suddenly anywhere, because it's the local people that pay most of them.
(And, no, this is not fair. And locals should push for higher pay.)
That is what's happening in Cuba where highly skilled people with good English skills rather work in tourism than as doctors, engineers or teachers. Not a good long-term situation.
And then Kenya doesn't gain the benefit of the $2/hr wages (which, as mentioned, is already above average), plus the benefits in knowledge & experience, improved infrastructure, and secondary jobs, as well as the possibility of similar jobs in the future.
China started out competing almost exclusively on wages (like Japan, Korea, and Taiwan before them), and now the US and EU worry about Chinese advanced industries (eg. semiconductors) surpassing their own, the same way Taiwan did with chips and Japan did with cars.
Competing via low wages is, unfortunately, a necessary first step to growing an economy.
How much does rice cost? Beans? Other subsistence food? If it's all cheaper than the US, then not pointing that out (or showing how a wage compares to the average for a country), it's misleading journalism.
Of course, PPP doesn't constrain itself to subsistence food.
Also, they were labeling graphic and illegal (in the US) content including child porn.
Seems like paying them around the minimum wage for a receptionist is way way too low to me.
Which makes $2 a RELATIVELY better wage, however for the work - basically content filtering through the worst of the internet - there should be a higher compensation and psychological help available.
In this case, if Facebook and other online services had it available, they would have been better off reusing what they already developed for their side of content filtering. I want to believe 99% of bad content is already automatically filtered out, and anything that comes through and flagged by content moderators fed back into the systems that automate the process.
> The data labelers employed by Sama on behalf of OpenAI were paid a take-home wage of between around $1.32 and $2 per hour depending on seniority and performance.
includes a section about Google hiring Americans to do vaguely similar work, at $15/hour. I say "vaguely" because I don't think my cousin Missy had much racist or sexual content to deal with; at least that wasn't the whole point of the job like the Sama example.
Some of you may have heard this story before. The big company uses a small company to be insulated from the legal complications of hiring. That figured into the story: Google claimed they couldn't fix an obvious inequity, and HR told them to try harder. Missy got paid.
(In case you're wondering, that's not a fake name. I showed this to her beforehand, and she said it was fine.)
I could pay a Kenyan $17 an hour, or I could pay a Kansan instead and get US government benefits. The same applies at every payment level. Your logic fails the moment you realize that not every person/worker on earth is equal.
Insisting on this sort of articles just goes to show how sheltered these offended people are. I live in the third world and I know people who are thankful to Coca Cola for their "$2 wages."
(And I think most of what you wrote is based on similar stereotypes)
What stereotypes?
That would be transformational.
To remain stable, economic growth must be slow and steady. The alternative is you simply don't go to Kenya, rather, you go somewhere else, and Kenyans get $0/hour.
I also didn’t suggest anything like paying way over market.
Also...yes it is a sentiment (a view of or attitude toward a situation or event; an opinion). Our opinions about what constitutes "taking advantage" are different. Saying otherwise doesn't make your argument more compelling.
I think the question we are all considering is why OpenAI behaves differently overseas than they might when trying to poach a smart engineer from a competitor in the Us.
This is a bad take. Globalization (what you call "taking advantage") raised a billion people out of abject poverty in the last few decades.
When a company pays more than the prevailing wage, they're doing good and deserve kudos. Expecting economic deals with less developed countries to be driven by charity (which is what you're advocating for) rather than mutual benefit is A) paternalistic and demeaning, and B) cannot possibly be the basis of sustainable economic gains for developing countries.
Doesn't it? If you have some kind of objective measure of fairness, I would like to hear it.
What kind of business model is one that relies on exploitation of extreme poverty in order to succeed?
that's a very good question, but asking it as if it is a problem of OpenAI and not a problem of literally all of our economy sounds a little strange to me.
That is just economic nonsense. Labor productivity of economy depends on utilization of capital and ideas in that economy and it is pretty hard to increase. Average wage per hour in that economy cannot be higher than average productivity per hour. If you set minimal wage higher than its productivity, then you just destroyed almost all legal employment and forced everyone to grey economy or subsistence agriculture.
I myself do work with mediation of content for several social networks, getting a little more than that. The only problem is that the monthly hours are irregular, there are very few hours.
If someone has an offer of 4 dollars an hour for 200 hours a week without any benefits, I accept it immediately. And that's half the US salary, as I understand it.
We’ve had schools changing policies, entire products being created to detect if a piece of work is AI-generated, countless articles on the impressiveness of the technology, even more articles arguing about some aspect of it.
And this was all on HN.
ChatGPT has been one of the most immediately picked up and heavily used and discussed technologies in the last few years. And it has permeated into the regular zeitgeist in a way that hasn’t happened in even longer. I don’t understand how you could even remotely say otherwise.
People have no way of knowing where the things they buy come from or how to source things ethically unless people write articles like this and push for higher standards when they see an opportunity.
It’s one thing to protect genuinely well-meaning people who can’t meet a purity standard, but OpenAI has a track record of dishonesty (starting with their name).
I don't know about other places and maybe I'm an outlier but I check everything for where it's produced / manufactured. Most if not all items in the US are required to have where the item was produced/manufactured. Some may say "Manufactured in the US with globally sourced material" makes it harder to pin down where each of the components originated from. Clothes and other items the country of origin is on the tag.
As a resident Indian, as far as I know (a) children have a guaranteed right to education which keeps them in school until 14 (b) children in fabric sweatshops was never an Indian thing. Children in brick kilns and beedi factories rolling beedis because of their nimble fingers used to be a thing.
India’s census figures are mostly reliable, unlikely to be off by an order of magnitude.
https://en.m.wikipedia.org/wiki/Child_labour_in_India
There are roughly 250 million child laborers globally of which 11 million are Indian children so around 4.4% of the global child labor population is Indian.
I.e., despite having 14% of the world's kids, only 4.4% are child laborers. I'd say India is moving in the right direction when it comes to child welfare.
https://www.culturalworld.org/what-percentage-of-the-worlds-...
None of the quoted stats or linked articles support the assertion that things are improving. The only trend that I can see noted is that Indian child labourers are moving to the cities to do manufacturing work and abandoning rural/agriculture work which on the face of it doesn't seem like a fully positive development development all.
OpenAI went shopping for Kenyan labor and those who did the job were not American and never interviewed in SF
Salary is a result of market.
A bay area worker may view this as exploitation, while I say "please god may this never end"
As another poster said- in Kenya a job is a job and they still paid above average. Money going to Kenya is a good thing. They are doing the work so it must be worth it.
> Almost more interesting in that article I found that Sama, the Kenyan company, was also asked to collect CP and brutal/gore images for OpenAI. After delivering 1400 images, Sama cancelled the contract as this was even too much for them. OpenAI then talked about "miscommunication" and they actually did not really want those CP images. Well, they made its own category for it and asked Sama to collect images of several other categories, but somehow it was lost in communication that for one of theses categories, no images should be collected. Because they are illegal. OpenAI swears they never opened the images they received (and paid for).
https://news.ycombinator.com/item?id=34427118
If someone agrees with your premise that "these days the human cost is just another cost", then it seems logical that a newspaper editor/author will shape the headline about economic suffering, instead of mental suffering.
By adding more information to the headline about "$2 per hour", it might make the discussion more confusing by much of the comments discussing the economic part. BUT, a clickbait-y title might increase the reach of the article.
"Kenyan workers helped make OpenAI's ChatGPT less toxic"
"OpenAI paid Kenyan workers minimum wage to work on ChatGPT"
Many different things are objectively true, but of course they mention the pay, because they know that to most of their "Western" readers it sounds outrageous.
Regardless, I should have included the phrase "depictions of" or similar.
If they're not properly novel then it depends on what the source images were, I guess? But it's not like you'd need to train it on anything obscene for it to figure out how to combine bad things with nudity.
2. It says OpenAI "used" workers which has a negative connotation.
3. It is factually incorrect: OpenAI paid a vendor $12.50 per hour. It's unclear if they knew how much the actual workers were paid.
4. It used a negative framing implying ChatGPT is toxic.
So something like
"OpenAI vendor paid Kenyan workers above median wage to make ChatGPT more safe"
But I too wonder what the cost would be as a comparison.
If quality developers could be hired for $2 per hour, that would be fantastic. Imagine how much progress we could make. The first thing I would do is hire a bunch of people to fix the bugs in my favorite open source software.
In the short term it might make my life harder as I would probably lose my high paying development job. But that's absolutely not a good reason to oppose progress.
Developer salaries even in such countries tend to be far better than $2/h, but that's explicitly because western companies like these "were hiring developers there instead", if they weren't, the developer salaries there would match the local market and be even lower.
But you can't escape that gap so long as you're fundamentally trading time for money.
If you want to capture more of the value of your work, you become a contractor and bid firm-fixed or cost-plus contracts instead of labor-hour or time-and-materials.
If you want to capture even more value than that, you work on building business systems and selling a valuable product or service rather than spending your work hours doing the tasks themselves.
Whether you're in Kenya or the US and whether you're doing undesirable or highly desirable work, it is not a moral failing of the business you are working for that they did not also take the time to ensure that you could have become a contractor if you wanted to or that they did not ensure that your economic situation allows sufficient market opportunities for you to quit and form a business if you wanted to.
Does this qualify as nominative determinism, given OpenAI founder Sam Altman's username is... sama?
Where did you earn your law degree?
For those looking for an AI data labeling service also tries to help workers along the way, here's a plug to the company I started "dataprep.be" [4]
We have a small preference for working workers with special needs (deaf, mute and employees with small handicaps). Public subsidies for these type of workers help our case in the EU as well as contraints for some public institutions in the EU to hire more handicap workers.
With clients more sensitive to costs, we also work with remote data labelers from developing countries. We help putting checks in place to limit forced and child labor. We pay 5% extra so they have time to learn high demand tech skills. (Using Khan Academy and free access to a normally 250$/year Datacamp subscription)
Happy to work with the HN crowd or just receive feedback and mentoring! (My email is in my profile)
[1] https://www.vice.com/en/article/88apnv/underpaid-workers-are... [2] https://gizmodo.com/horror-stories-from-inside-amazons-mecha... [3] https://www.thebureauinvestigates.com/stories/2022-10-20/beh... [4] https://www.dataprep.be
Surely 3 years is too little time to be forgotten.
I suppose none of us gets to go to Kislovodsk ..
https://www.youtube.com/watch?v=XrAGf2EMAbs
edit: This is mentioned in the original article. From TFA:
>Sama delivered OpenAI a sample batch of 1,400 images. Some of those images were categorized as “C4”—OpenAI’s internal label denoting child sexual abuse—according to the document. Also included in the batch were “C3” images (including bestiality, rape, and sexual slavery,) and “V3” images depicting graphic detail of death, violence or serious physical injury, according to the billing document. OpenAI paid Sama a total of $787.50 for collecting the images, the document shows.
> Within weeks, Sama had canceled all its work for OpenAI—eight months earlier than agreed in the contracts. The outsourcing company said in a statement that its agreement to collect images for OpenAI did not include any reference to illegal content, and it was only after the work had begun that OpenAI sent “additional instructions” referring to “some illegal categories.” “The East Africa team raised concerns to our executives right away. Sama immediately ended the image classification pilot and gave notice that we would cancel all remaining [projects] with OpenAI,” a Sama spokesperson said. “The individuals working with the client did not vet the request through the proper channels. After a review of the situation, individuals were terminated and new sales vetting policies and guardrails were put in place.”
Be careful here. Read the article closely, here is what it says:
> [Sama] said in a statement that its agreement to collect images for OpenAI did not include any reference to illegal content, and it was only after the work had begun that OpenAI sent “additional instructions” referring to “some illegal categories.”
Note how carefully this is worded - if Time could confidently say that OpenAI asked for C4 images, they would have absolutely put that in the article. Now, this is filtered through PR statements, but it reads to me like a poorly-worded email went out from OpenAI that didn't actually ask for C4 images, but one of Sama's employees interpreted it as an ask, and started collecting without raising any red flags up the chain. And they got fired for it.
Also, I'm not up on the laws regarding this stuff, but are the other, awful categories illegal to collect? If not, there's not much room for ambiguity.
All your word salad ignores this reality.
But I see why "serious business" investors will want to avoid anything controversial. It is their business to suck life out of everything.
That makes it vulnerable to disruption.
I'd love to know how much of its seeming smart is because it's just that good and how much is because humans directly intervene, on some level, in how it responds. Is it that much better than other efforts, purely from machine learning, or has it just had tons and tons more human hours put into tweaking it than others have? How much of it's "real" AI and how much is good ol' hand-written decision trees?
I asked to write a 100-words first message for a Tinder match who had "very talkative" in their bio. ChatGPT kept saying how it would be rude of me to send such message and both the other person. FFS.
I confess I feel silly not even considering it according to your surprise. Had a whole wave of thought about the market of facts. Which "truths" are you most attached to and how much are you willing to pay? Do you want Jesus to be our savior or not? What's it worth to you?
https://time.com/6147458/facebook-africa-content-moderation-...
Uber did Greyball and nothing happened. Its corporate charter should've been revoked and the executives jailed. A very clear message has been sent to the tech companies by this: do whatever you want, the law doesn't apply to you.
OpenAI did this and nothing happened.
At the next pandemic, some sort of this generative AI will be used to write the anti scientific articles and woe upon to you then. This time it was bad but that will be a thousand times worse.
Can't find who said OpenAI basically released a zero day on human cognition without a patch. Outlaw it already until we find one. This is no joke here.
The next pandemic when these articles will be written by these AI will be a lot worse. You can write a thousand very scientific sounding papers in seconds.
Even right now to appease these morons the world has largely dropped masking mandates and that kills and disables people. There is no scientific reason not to mask and every reason to do it but it's politically not expedient. Check https://jessicawildfire.substack.com/p/heres-150-sources-on-...
So the danger is the death of millions in the next pandemic. And that's just what immediately comes to mind as the biggest, most concrete danger.
Other dangers include the almost immediate obliteration of independent artists -- there's a lawsuit now, good but it won't be able to do much, even a few hundred millions in damages is just cost of doing business to these folks. So many possibilities... https://mobile.twitter.com/Dereklowe/status/1599035870308618...
I think that vastly overestimates the incremental value of more articles. Despite the common refrain, most people are not "doing their own research" and reading articles, so whether we have more real or fake ones is not really that material.
> Other dangers include the almost immediate obliteration of independent artists -- there's a lawsuit now, good but it won't be able to do much, even a few hundred millions in damages is just cost of doing business to these folks. So many possibilities... https://mobile.twitter.com/Dereklowe/status/1599035870308618...
Computers obliterated millions of jobs as well. Maybe we should outlaw those as well?
> Even right now to appease these morons
As a meta point, your barely veiled loathing of "these morons" makes your protestations that they must be protected from these technologies for their own good pretty ironic.
Governments trying to track down and stop people from being their own taxi driver, how ridiculous. Why was Uber not charged? Because their own customers called their reps and got the laws changed. Wouldn't have happened the other way around.
I don't know why you luddites comment on a tech forum trying to ban everything you don't like or that sounds scary. Is the big bad bitcoin going to bite you while you sleep? Did the AI burn your toast?
There is a section that mentioned "An online search for the text reveals that it originated from an online erotica site". If you dont want sexually explicit text, then maybe dont include text from literary erotica sites in the first place? Or is it like advertising where companies keep passing around the same chunk of data and continue adding to it? Still, I would think making or paying for better(smarter) scrapping bots would be cheaper.
At this point, it's cheaper for them to manually alter the output than it is to retrain on a more appropriate data set.
OpenAi paid Sama $12/hr for the work and they in turn paid their employees $2/hr.
Yet because workers were reportedly affected by the content Sama cancelled the contract?
I think we all know they would just replace the $2/hr workers.
It doesn’t check out
This kind of shit really boils my piss.
I did extra work on the Alexander movie in Morocco. There was a pay disparity between what the local production company were paying Westerners vs what they paid locals [1]. I'm almost 100% sure the production company paying the local company would have been paying the same price for both.
[1] - this isn't even the half of what was wrong with the situation; they were filming during Ramadan, so the locals couldn't eat/drink/smoke anything during the daylight hours and made to stand around in the blazing sun. One old lady fainted. The production company asked the Westerners to please try and keep out of sight whilst drinking/eating/smoking between takes, but then at the same time telling us we can't move from our spots lest we break continuity. Towards the end of filming it almost turned into a riot due to the bad treatment of the locals. Total joke.
Sama could have provided the mental health support that employees may have needed with the 10$/hr per employee profit they were making. Sama didn't and decided that instead of losing profit by providing their employees with mental healthcare they canceled the contract.
Smells fishy.
That's sufficiently risky for a company to be prepared to fire a high profile client, and whilst their margins were huge the value of the actual project was only $200k, which isn't going to seem like riches to Sama's California-based management team.
It's unproductive to talk about the hundreds of billions of dollars of slavery work that companies do until we have an answer for the trillions of dollars of toxic debt created by these huge organizations.
What are the other countries?
I know China is doling out these loans apart of their belt road initiative but I'm not up to speed on other countries hamstringing countries with unneeded infrastructure and billions in loans.
Alex Gladstein is writing a new book about this huge problem, as the latest good books about it he found were in the 70s.
I live in a similar country (Nigeria)…We Africans are the source of our own problems because we choose not to utilize our resources well and plunder the few we extract, not some boogeyman IMF that also gave loans to some countries that are prosperous today.
In return he increases income taxes, devalues the money and sets up the country for exporting luxury goods that are needed for 1st world countries (but local people would have no chance of buying) why destroying their own agriculture, so that they have to import inferior mass manufactured products from rich countries. On the long term the countries have to give back 10-20 times the original loan that is gone in a few years.
Does any of this sound familiar to you?
"To get those labels, OpenAI sent tens of thousands of snippets of text to an outsourcing firm in Kenya, beginning in November 2021. Much of that text appeared to have been pulled from the darkest recesses of the internet. Some of it described situations in graphic detail like child sexual abuse, bestiality, murder, suicide, torture, self harm, and incest."
[1]https://www.calderdale.gov.uk/wtw/search/controlservlet?Page...
[2]https://api.parliament.uk/historic-hansard/commons/1842/jun/...
OpenAI hired Kenyans because of limited labour law, no unions, no red tape, and the cheapest they can find.
Getting these jobs is good for Kenya
For simple outsourcing tasks like this one, outsourcing to Kenya is a win-win. You can drastically reduce costs, while significantly improving people's lives.
That'd be true regardless of labor laws, unions, or otherwise. The legal system in Kenya isn't quiet as wonderful for business as you describe; it inherits British colonial bureaucracy, has some corruption, isn't all that laissez-faire, and can be a minor pain in the butt. Unions, likewise, would increase income, but not by enough to make this a bad deal.
Kenya is competing with India and the Philippians here, not with low-cost US labor.
Are all the sweatshop and slave labor[1] jobs created by Zara good for those children and slaves? You get cheap clothes out of it after all. But is it really helping anyone other than Zara, and their CEO who is worth ..checks notes... 50+ billion dollars?
Companies could help a lot more by increasing labor standards, helping those in need, paying above a living day to day wage, etc. The would still have a few billion left in the bank I believe...
[1]https://www.forbes.com/sites/andersonantunes/2011/08/17/zara...
Well, you could ask people from Kenya what they think about this stuff, instead of speaking on their behalf.
I believe someone from there even commented on this, in the thread.
>Agents, the most junior data labelers who made up the majority of the three teams, were paid a basic salary of 21,000 Kenyan shillings ($170) per month
Based on my quick and dirty googling, it looks like this is an above average salary for Nairobi.
>Are all the sweatshop and slave labor[1] jobs created by Zara good for those children and slaves? You get cheap clothes out of it after all. But is it really helping anyone other than Zara, and their CEO who is worth ..checks notes... 50+ billion dollars?
You haven't established a connection between OpenAI and slave labour. So this is irrelevant and detracts from your point by costing you credibility.
> If this signals to other businesses that Kenya is a great place for cheap labor and little regulation, it could very easily be turn bad for many Kenyans.
That's for Kenyans to decide.
It's awful but who sends these kids to work? What will a mother of 3 children do if she can't feed herself?
It's a terrible cycle but it takes ages to fix. No foreign company is benefitting from Bangladeshi kids making low-quality bricks for their village. What is the solution today? If you tell them they can't work, who will feed them today?
Are you personally going to donate your money to not only feed all these people, build schools and infrastructure, but also $20/hour to everyone involved in the process?
Same arguments were used for slave labor. How are they going to take care of themselves? Might as well just keep them as slaves. What are we going to do, pay them?
Child labor has been used by just about every major fashion brand in the world, not just to "make bricks for their village"... If that's the extent of what you think child labor is, might want to check out some books or do some googling. Here's just one to get you started: [2].
[1]https://www.forbes.com/sites/andersonantunes/2011/08/17/zara... [2]https://en.wikipedia.org/wiki/Nike_sweatshops
You see poor people and are so disgusted by them and their poverty, you blame the people who do pay them rather than those who don't. How many people could employ a Kenyan today, but don't?
How backwards is this thinking?
Imagine if every American company outsourced 5% of its back office work to Kenya for $2. That would be an enormous benefit, think of the millions of lives improved. Every extra foreign dollar paid to Africans working remote service jobs has 100x the impact of paying it to an American.
Do you really think it's more moral to increase the pay of American office workers from $22 to $24, rather than adding another Kenyan for $2?
You would pay the Americans because they have unions? labor laws? red tape? WHO GIVES A SHIT. "Sorry Kenyans, come back when you pass a minimum wage of $15 USD, then we'll employ you. Oh you can't meet a $15/hr productivity? Stay poor then".
Yes ChatGPT needs to be “not toxic”.
The company leadership believes this and is investing heavily researching and designing and implementing as many safe guards as possible.
They know if they don’t do a good job here it will be a PR and support disaster and ultimate effect the bottom line.
No serious customers will buy any AI product if there’s a chance it spews sex, hate speech or violence, etc. to their end users.
At least for now, some humans somewhere in the chain will have to review toxic content to help train.
It sounds particularly depressing to me that a $50B company can outsource this to Kenya for $200k. But I don’t know who else or how much money would make it better.
Totally hypothetical but I bet some of the handsomely paid US engineers have a ton of policies in place to eliminate exposure to toxic content. But also when they do see some they have a much better foundation in life for it not triggering major mental health problems.
What we really need to know is what OpenAIs whole portfolio of approaches, budgets, staff and outsourcing looks like.
In the meantime kudos to Time for exposing one piece for public discourse.
In other words, you want OpenAI to be...open? :)