Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Anyone else witnessing a panic inside NLP orgs of big tech companies?

old.reddit.com

Post reply on HN
Print view View original

221–230 of 527 posts

Re: Anyone else witnessing a panic inside NLP orgs of big tech companies?

#221

Post by screye »

hn_throwaway_99 wrote: post #119

Wow - this is just wild. I've seen lots of arguments around "AI won't take everyone's job, it will just open up new areas for new jobs." Even if you take that with the benefit of the doubt (which I don't really think is warranted): 1. You don't need to take everyone's job. You just need to take a shitload of people's jobs. I think a lot of our current sociological problems, problems associated with wealth inequality,…

> human translators, medical transcriptionists, court reporters

Yes, they will be all called 'ai data labellers'.

For a long time, "People don't just want jobs, they want good jobs" was the slogan of industries that automated the boring stuff. Now AI is suddenly good at all the jobs people actually want and the only thing it can't do is self-improve. In an AI future, mediocre anything will not exist anymore.

Either you are brilliant enough to be sampling from 'out of distribution', or you're in the other 99 percent normies that follow the standard : "learn -> imitate -> internalize -> practice" cycle. That other 99% is now and eternally inferior to an AI.

Re: Anyone else witnessing a panic inside NLP orgs of big tech companies?

#222

Post by elwell »

timoth3y wrote: post #202
hn_throwaway_99 wrote: post #119

Wow - this is just wild. I've seen lots of arguments around "AI won't take everyone's job, it will just open up new areas for new jobs." Even if you take that with the benefit of the doubt (which I don't really think is warranted): 1. You don't need to take everyone's job. You just need to take a shitload of people's jobs. I think a lot of our current sociological problems, problems associated with wealth inequality,…

The problem is not that automation will eliminate our jobs. The problem is that we have created an economy where that is a bad thing.

The problem is that humans are often selfish.

Re: Anyone else witnessing a panic inside NLP orgs of big tech companies?

#223

Post by shawntan »

iandanforth wrote: post #159
swatcoder wrote: post #117

Earlier quoted context omitted.

> There is no good reason this should not work everywhere else, in exactly the same way. Take for example a large retailer who has a large internal knowledge base. Train an LLM on that corpus, ask the knowledge base any question. Since LLM’s can’t scope themselves to be strictly true or accurate, there are indeed good reasons, like liability for false claims and added traditional support burden from incorrect guidanc…

"LLM’s can’t scope themselves to be strictly true or accurate" This isn't true though the techniques to do so are 1. Not as yet widespread 2. Decrease the generality of the model and its perceived effectiveness.

I'm interested to hear what these techniques are. Decreasing the generality will help, but I fail to see how that scopes the output. At best that mitigates the errors to an extent.

Re: Anyone else witnessing a panic inside NLP orgs of big tech companies?

#224

Post by yieldcrv »

siva7 wrote: post #86
davidkuennen wrote: post #38

Earlier quoted context omitted.

The most amazing thing about ChatGPT translation is, that you can even instruct it how to translate. For example "dutzen" and "sietzen" in German. I just simply tell it how it should do it and it did. Absolutely amazing. It's like actually working with a real translator.

That's something i'm really sorry for but those jobs will be likely the first to fade away, there is a whole university faculty dedicated to the profession of the language translator where i live.

goodbye to teaching English in Asia! come on home ya'll!

Re: Anyone else witnessing a panic inside NLP orgs of big tech companies?

#225

Post by pcthrowaway »

timoth3y wrote: post #202
hn_throwaway_99 wrote: post #119

Wow - this is just wild. I've seen lots of arguments around "AI won't take everyone's job, it will just open up new areas for new jobs." Even if you take that with the benefit of the doubt (which I don't really think is warranted): 1. You don't need to take everyone's job. You just need to take a shitload of people's jobs. I think a lot of our current sociological problems, problems associated with wealth inequality,…

The problem is not that automation will eliminate our jobs. The problem is that we have created an economy where that is a bad thing.

I don't think it's the economy, it's the policy. Automating a shit-ton of jobs is great for the economy. The economy is just fine if 90% of people are starving because big corps are saving shit-tons of money.

The government of a wealthy country should ensure that its citizens are able to eat, and have a sheltered place to sleep, without them needing to work. Because the way things are going, there won't be enough work to go around. Even now, with the supposed "labour shortage" there are record numbers of homeless people, and people living paycheck-to-paycheck. Housing is more unaffordable than ever. Minimum wage is not keeping up with the economic realities.

Governments need to step in; they need to change policy so big corps are paying more taxes, and that tax goes to a basic income that can cover the cost of housing and the cost of food. Maybe not right away, maybe it starts at $100/month. But eventually the goal should be to get everyone on a basic income that can cover the necessities, then if they want to be able to enjoy luxuries (concerts, gourmet food, hobbies, streaming services, etc.) they can choose to work.

Re: Anyone else witnessing a panic inside NLP orgs of big tech companies?

#226

Post by pyuser583 »

hn_throwaway_99 wrote: post #119

Wow - this is just wild. I've seen lots of arguments around "AI won't take everyone's job, it will just open up new areas for new jobs." Even if you take that with the benefit of the doubt (which I don't really think is warranted): 1. You don't need to take everyone's job. You just need to take a shitload of people's jobs. I think a lot of our current sociological problems, problems associated with wealth inequality,…

The problem starts long before AI takes the jobs.

I used to do a job that was eventually automated. We did the one and only thing the computer couldn’t do - again and again in a very mechanical fashion.

It was a shit job. You might get promoted to supervisor - but that was like being a supervisor at McDonalds.

Why not treat the job seriously? Why didn’t the company use it as a way to recruit talent? Why didn’t the workers unionize?

Because we all knew it would be automated anyway.

We were treated like robots, and we treated the org like it was run by robots.

There’s a huge shadow over the economy that treats most new jobs like shit jobs.

Re: Anyone else witnessing a panic inside NLP orgs of big tech companies?

#227

Post by aleph_minus_one »

xp84 wrote: post #149
csa wrote: post #135

Earlier quoted context omitted.

> I think a lot of our current sociological problems, problems associated with wealth inequality, etc., I see where you’re coming from, but is this really the main source of the inequality? Based on numbers relating to workers’ diminishing share of profits, it seems to be that the capital class has been able to take a bigger piece of the profit pie without sharing. In the past, companies have shared profits more wide…

I think you and the one you're replying to are both very right. Yes, more of this money is going, instead of middle-class workers, straight to the capital class who own the "machines" that do the work people used to do. Except instead of it being a factory that makes industrial machines owned by some wealthy industrialist, the machines are things like Google and AWS and the owners are the small number of people with…

> It's really striking though that a person graduating high school in say, 1970, could easily pick from a number of career choices even without doing college or even learning an in-demand trade, like plumbing, welding, etc. [...] Those are the days that are super gone

Isn't this rather a strong argument for the claim that what high school as of today teaches is a strong mismatch with what the labour market demands? In other words: the pupils are taught skills for many years of their life that are rather worthless for the job market.

Re: Anyone else witnessing a panic inside NLP orgs of big tech companies?

#228

Post by WalterBright »

1attice wrote: post #204
WalterBright wrote: post #161

Earlier quoted context omitted.

Imagine the devastation wrought by automatic looms, that put all the weavers out of a job! 97% of jobs used to be working on the farm. Now it's something like 2%.

This hoary take irks me. There were still places for human endeavour to go when the looms were automated. That is no longer the case. Think of it instead as cognitive habitat. Sure, there has been habitat loss in the past, but those losses have been offset by habitat gains elsewhere. This time, I don't see anywhere for habitat gains to come, and I see a massive, enormous, looming (ha!) cognitive habitat loss. -- EDIT…

> There were still places for human endeavour to go when the looms were automated.

Your job didn't exist then. Mine didn't, either.

Re: Anyone else witnessing a panic inside NLP orgs of big tech companies?

#229

Post by hn_throwaway_99 »

jMyles wrote: post #192
hn_throwaway_99 wrote: post #119

Wow - this is just wild. I've seen lots of arguments around "AI won't take everyone's job, it will just open up new areas for new jobs." Even if you take that with the benefit of the doubt (which I don't really think is warranted): 1. You don't need to take everyone's job. You just need to take a shitload of people's jobs. I think a lot of our current sociological problems, problems associated with wealth inequality,…

> given our current economic systems What can possibly be the benefit of requiring this constraint? Remove the idea that this is necessary and watch how much relaxation comes to the deliberation on this topic. "Current economic systems" will simply have to yield. Along with states. This has been obvious for decades now. Deep breaths, everybody. :-)

> What can possibly be the benefit of requiring this constraint?

It's not "requiring this constraint". If you have some plausible pathway to get from our current system to some "Star Trek-like nirvana", I'm all ears. Hand-wavy-ness doesn't cut it.

> "Current economic systems" will simply have to yield.

Why? For most of human history there were a few overloads and everyone else was starving half the time. Even look at now. I'm guessing you probably live a decent existence in a decent country, but meanwhile billions of people around the world (who can't compete skills-wise with upper income countries) barely eke out an existence.

For the world that just lived through the pandemic, do you honestly see systems changing when worldwide cooperation and benevolence is a prerequisite?

Re: Anyone else witnessing a panic inside NLP orgs of big tech companies?

#230

Post by shawntan »

mr_toad wrote: post #213
dongobread wrote: post #107

I worked in a research capacity in the voice assistant org of a big tech company until very recently. There was a lot of panic when ChatGPT came out, as it became clear that the vast bulk of the org's modeling work and research essentially had no future. I feel bad for some of my colleagues who were really specialized in specific NLP technology niches (e.g. building NLU ontologies) which have been made totally obsole…

“ Seeking an improvement that makes a difference in the shorter term, researchers seek to leverage their human knowledge of the domain, but the only thing that matters in the long run is the leveraging of computation. “ http://www.incompleteideas.net/IncIdeas/BitterLesson.html

I've seen many interpretations of this article and I'm curious as to the mainstream CS reading of it.

One could look at the move from linear models to non-linear models or the use of ConvNets (yes I know ViTs exist, to my knowledge the base layers are still convolution layers) as 'leveraging human knowledge'. Only after those shifts were made did the leveraging of computation help. It would seem to me that the naive reading of that quote only rings true between breakthroughs.

Post reply on HN