> In a snake-eating-its-own-tail irony, a 2023 analysis found that between 33% and 46% of workers on the platform were using large language models to complete their tasks,
I assume AI use by workers has risen to the point where it renders Mechanical Turk pointless.
Yeah, I was doing this kind of Artificial Artificial Artificial Intelligence back in 2012 to make some extra $$$. Glad they finally "patched" that hole ^^.
That's just Artificial Artificial Intelligence, the triple negative implies they built an automated system to impersonate humans who impersonate an automated system (which ultimately imitates a human).
Artificial Artificial Artificial intelligence is when the chat bot is out of capacity, so a person in India is writing the response that gets returned by the LLM which gets pasted into Mechanical Turk.
Are they? If one uses artificial in the sense of "fake" then a human pretending to be a machine AI would count as an artificial AI. The only scenario where this doesn't hold is if you are using "artificial" in the sense of "not created by nature".
This likely means those consuming the outputs of Mechanical Turk don't have a good way to measure the value (aka quality) of the outputs.
If they did - then they shouldn't care whether it's a human or a LLM. And if it's a LLM - then the cost will roughly correlate to the MIN(cost of the LLM, cost of a human) to do the task.
I think the "state of the art" of measuring the quality of outputs was to send the same task to multiple "agents" and only accept answers if over a certain amount agree. With some human review and reputation scoring sprinkled on top. It was a while since I was in this field though
This approach does work when there's a clear answer but what about tasks where the correct answer is multi-modal? Incentivizing agreement works only for tasks where there's clear answers.
The problem is bigger. Outside of coding, there is no real way to reinforce a model with pass/fail cycles until it stops hallucinating. This is why customer service uses will always have a problem. This compounds as you chain agents together.
It's like the speed of light - to get to that point, you need exponentially more energy, and you will never ever get there.
> then they shouldn't care whether it's a human or a LLM.
I imagine that the whole point of posting a task to Mechanical Turk nowadays is that you want it to be completed by humans. Either because you are after the small discrepancy between AI and human performance, or because humans are the object of your investigation.
If you get someone to mow your lawn, do you not care if (a) they use an automated machine that you could rent by the hour for far less, and/or (b) at the end your lawn is actually in a good shape?
I have very low expectations for $2 per hour. If I would feel the need to use such service I would expect them to cut all corners. If you have a subscription for a decent LLM, know how to get my data into it and know how to write a prompt that does the job you might have 3 skills I don't or I might not have time to do it.
If the lawn looks great you get to do it again. I could ask the guy you send how much you are paying him and I could ask you how much the mower costs but it wouldn't necessarily be a good use of my time.
Say you notice they come in the morning, put a small machine on the grass, hit a button, and leave and come back 4h later to pick up the machine and charge you - wouldn't you be at least curious to see how much this machine costs? Or if there was another company that would lease such machines for much less than what you paid per hour, wouldn't you possibly prefer to contract them, even if you may have to pick up and drop off the auto lawn mower yourself?
And conversely, say that the person or machine that you paid to mow the lawn instead drenches your yard in herbicide and then salts the earth. Or vice versa, it starts planting weed seeds, or brings a bunch of ticks in the grass etc. You would probably not be happy to let this process finish and then say "oh well, guess I won't be hiring them again!" and just move on.
The whole point of delegation is to free up time to do something else. There is a store here where the owner always runs around like a madman. I ask him why he doesn't have his employees do all those things (they mostly stand idle chatting) He said: Then I have to explain everything and they will probably do it wrong.
If I'm getting a $200 mowing bill it could be worth looking at the machines. If I pay you $4 per week I wouldn't worry about it.
Back around 10 years ago, I gained a new manager who had previously managed mechanical turk. It was already recognised as a dead end back then.
I remember him talking about getting a mandate from Amazon Security to upgrade from the long EOL MySQL 4.0 to MySQL 5.something, and that it was almost impossible to get any resources committed from leadership to even do it despite the fact it was security requiring it (which usually resulted in everyone jumping before stopping to ask how high to jump). I want to say he ended up doing it himself? Something like that..
All existing extremely minimal headcount was tied up in a massive technical debt of KTLO work, and proposals to resolve those issues similarly met resourcing road-blocks.
Maybe the most unambiguous "ai will automate work" example I've seen yet.
Absolutely does not imply the workers are automated since they can now use the current models to do more complex tasks at the vast number of new AI training data startups.
Turk was simply not designed for greater complexity tasks and so much of their lunch has been eaten by startups specifically built to collect AI training data.
I turked for a bit trying to make some extra cash leading up to my wedding, but it was a very time-inefficient way to make money. I think I managed to wring 10 or 20 bucks out of it tops after plugging at it for a month.
I can see a high value startup, that will provide Human Intelligence with real Humans, locked in the room, with no network, books, LLMs and monitored 24x7 with cameras.
This has very little to do with “AI replacing jobs” and much much more to do with a bad product getting obsoleted by better ones.
Human labeling is a two sided marketplace and so as any marketplace startup knows, both sides need to be constantly nurtured otherwise the system can collapse as worsening quality leads to churn and a vicious cycle that empties out the platform.
In labeling, you need to understand the limitations of individual work and fatigue, keep your pipeline bursting with awesome and consistent work, and improve the platform to make customer experience great.
AMT has been totally languishing in all these respects. Pay is terrible, dishonesty rampant, etc. It was a bad product, no need to pedestalize it or turn it political
> In a snake-eating-its-own-tail irony, a 2023 analysis found that between 33% and 46% of workers on the platform were using large language models to complete their tasks,
I assume AI use by workers has risen to the point where it renders Mechanical Turk pointless.
Yeah, I was doing this kind of Artificial Artificial Artificial Intelligence back in 2012 to make some extra $$$. Glad they finally "patched" that hole ^^.
You were using LLMs in 2012?
They were faking artificial intelligence by using real individuals.
That's just Artificial Artificial Intelligence, the triple negative implies they built an automated system to impersonate humans who impersonate an automated system (which ultimately imitates a human).
Fiverr-5.5 was the leading model back then.
Not LLMs. (Useful) LLMs came to the market around 2022.
Artificial AI = stuff like mechanical turk where they get humans to do stuff computers can't do and make it look like it's "AI"
Artificial Artificial Intelligence = using computers to do mechanical turk jobs
You wrote the same thing twice, hehe.
But the point gets across.
Artificial Artificial Artificial intelligence is when the chat bot is out of capacity, so a person in India is writing the response that gets returned by the LLM which gets pasted into Mechanical Turk.
"Artificial AI"
and
"Artificial Artificial intelligence"
are the same thing.
Come on. This site can do better.
yes, this site can do a lot better than nitpicking acronyms
You can take that to the Automatic ATM Machine.
You're the one wasting time with another incorrect comment.
This point is that AAAI and AAI are not the same thing.
Are they? If one uses artificial in the sense of "fake" then a human pretending to be a machine AI would count as an artificial AI. The only scenario where this doesn't hold is if you are using "artificial" in the sense of "not created by nature".
The point is AI expands to Artificial Inteligence and Artificial AI expands to Artificial Artificial Inteligence.
Ya the second one should be Artificial Artificial AI. So close lol
If it wasn't made in the State of Artifice, then it's only Sparkling Intelligence.
Living, Low-income Minions?
This is my main argument as to why (people with) AI will not take over the world.
Cheap, disposable, on-demand intelligence has existed for millennia.
If anything, AI is more of an equalizer.
This likely means those consuming the outputs of Mechanical Turk don't have a good way to measure the value (aka quality) of the outputs.
If they did - then they shouldn't care whether it's a human or a LLM. And if it's a LLM - then the cost will roughly correlate to the MIN(cost of the LLM, cost of a human) to do the task.
I think the "state of the art" of measuring the quality of outputs was to send the same task to multiple "agents" and only accept answers if over a certain amount agree. With some human review and reputation scoring sprinkled on top. It was a while since I was in this field though
This approach does work when there's a clear answer but what about tasks where the correct answer is multi-modal? Incentivizing agreement works only for tasks where there's clear answers.
The problem is bigger. Outside of coding, there is no real way to reinforce a model with pass/fail cycles until it stops hallucinating. This is why customer service uses will always have a problem. This compounds as you chain agents together.
It's like the speed of light - to get to that point, you need exponentially more energy, and you will never ever get there.
> then they shouldn't care whether it's a human or a LLM.
I imagine that the whole point of posting a task to Mechanical Turk nowadays is that you want it to be completed by humans. Either because you are after the small discrepancy between AI and human performance, or because humans are the object of your investigation.
I don't see why I would care how they do the job. Just do the job, I have other things to do.
If you get someone to mow your lawn, do you not care if (a) they use an automated machine that you could rent by the hour for far less, and/or (b) at the end your lawn is actually in a good shape?
I have very low expectations for $2 per hour. If I would feel the need to use such service I would expect them to cut all corners. If you have a subscription for a decent LLM, know how to get my data into it and know how to write a prompt that does the job you might have 3 skills I don't or I might not have time to do it.
If the lawn looks great you get to do it again. I could ask the guy you send how much you are paying him and I could ask you how much the mower costs but it wouldn't necessarily be a good use of my time.
Say you notice they come in the morning, put a small machine on the grass, hit a button, and leave and come back 4h later to pick up the machine and charge you - wouldn't you be at least curious to see how much this machine costs? Or if there was another company that would lease such machines for much less than what you paid per hour, wouldn't you possibly prefer to contract them, even if you may have to pick up and drop off the auto lawn mower yourself?
And conversely, say that the person or machine that you paid to mow the lawn instead drenches your yard in herbicide and then salts the earth. Or vice versa, it starts planting weed seeds, or brings a bunch of ticks in the grass etc. You would probably not be happy to let this process finish and then say "oh well, guess I won't be hiring them again!" and just move on.
The whole point of delegation is to free up time to do something else. There is a store here where the owner always runs around like a madman. I ask him why he doesn't have his employees do all those things (they mostly stand idle chatting) He said: Then I have to explain everything and they will probably do it wrong.
If I'm getting a $200 mowing bill it could be worth looking at the machines. If I pay you $4 per week I wouldn't worry about it.
If you want data to train a Llm, you might want human data for specific parts vs Llm.
I guess that’s the dead internet theory. People are trying to train llms using mt, but just training on itself.
Back around 10 years ago, I gained a new manager who had previously managed mechanical turk. It was already recognised as a dead end back then.
I remember him talking about getting a mandate from Amazon Security to upgrade from the long EOL MySQL 4.0 to MySQL 5.something, and that it was almost impossible to get any resources committed from leadership to even do it despite the fact it was security requiring it (which usually resulted in everyone jumping before stopping to ask how high to jump). I want to say he ended up doing it himself? Something like that..
All existing extremely minimal headcount was tied up in a massive technical debt of KTLO work, and proposals to resolve those issues similarly met resourcing road-blocks.
Maybe the most unambiguous "ai will automate work" example I've seen yet.
Absolutely does not imply the workers are automated since they can now use the current models to do more complex tasks at the vast number of new AI training data startups.
Turk was simply not designed for greater complexity tasks and so much of their lunch has been eaten by startups specifically built to collect AI training data.
I turked for a bit trying to make some extra cash leading up to my wedding, but it was a very time-inefficient way to make money. I think I managed to wring 10 or 20 bucks out of it tops after plugging at it for a month.
I can see a high value startup, that will provide Human Intelligence with real Humans, locked in the room, with no network, books, LLMs and monitored 24x7 with cameras.
Please enjoy each task equally
24/7 isolation with no stimulation outside of work? Wonder if the hallucination rate would be higher or lower
Just give them some exercise bikes to pedal to keep them physically occupied
I'd suggest first looking into the conditions that enable humans to generate sustained, high quality output.
I'd pay for authentic artisnal fresh prison-to-screen Trump Truths produced that way.
Prison labor for the rescue!
I had no idea this was still around.
It helped me buy a Battlefield 2 "Special Forces" expansion pack back in the day.
Well, I could've bought it either way but buying it didn't impact my normal income because I did Mechanical Turk in my free time enough to get it.
That was a fun DLC. The zip lines were my favorite on the night maps.
This has very little to do with “AI replacing jobs” and much much more to do with a bad product getting obsoleted by better ones.
Human labeling is a two sided marketplace and so as any marketplace startup knows, both sides need to be constantly nurtured otherwise the system can collapse as worsening quality leads to churn and a vicious cycle that empties out the platform.
In labeling, you need to understand the limitations of individual work and fatigue, keep your pipeline bursting with awesome and consistent work, and improve the platform to make customer experience great.
AMT has been totally languishing in all these respects. Pay is terrible, dishonesty rampant, etc. It was a bad product, no need to pedestalize it or turn it political
Pour one out for the original A.I. (Actual Indians).
Where do I find participants for my user studies then?
https://prolific.com
It's a shame. Mechanical Turk works better than any AI.
It's still AI, just a different type.
The Actually Indian kind?
Hey, maybe they're Indonesian!
I thought they were Turkish.
I thought they only played Chess.
Ever been to Singapore? Their apartments have a room for a Indonesian maid.
Never underestimate just how cheap human life is!
mild racism, needs to be reported
It's a joke
you can call the police
India is not a race, therefore this is not racist.
Oh my goooood, the average Asian person eats more rice than they do bread.
It's not racism if you think Indians can do a better job than artificial intelligence.
American Idiot kind
Artificial Insemenation?
personally, i've never had good luck with MT's quality.
They moved all the Mechanical Turk workers over to robot and autotaxi piloting.
And monitoring of “cashier-free” grocery stores.