Your use case are way too general haha. They include many key things that LLMs currently don’t do. Anyway, maybe you’re not super interested in discussing whether they’re “Enough to usher in a new industrial revolution even without further progress.”, but if you were my next question would be whether the Internet would count as a new industrial revolution in your eyes. (I would say “no, but kinda / almost”, and I would say that the no LLM --> LLM transition looks like it’s kinda comparable-ish to the Internet transition.)
(I’ve been trying a new drug and my brain isn’t at 100% capacity, hence slow or limited replies right now.)
I think that’s a good question. I think the Internet doesn’t feel to me like it reorganized enough of how civilization works to quite be a revolution. In contrast to things like agriculture or steam engines where the vocation and living situation of so much of the population changed. I think LLMs, via automation, can cause an economic reorganization on the scale of agriculture/industrialization, that the Internet itself didn’t do. I’m fuzzier on where “electricity” fits.
I think LLMs, via automation, can cause an economic reorganization on the scale of agriculture/industrialization
But like, how specifically? I agree that there’s some idea around making a bunch of software significantly more beginner-friendly by giving it an LLM interface, and in some ways significantly more powerful with LLM “agents”. Is that a sufficient class of thing for what you’re referring to? I mean, do you think that 50% of people will be working on something different within 5 years, or something like that? Which 50%?
“Beginner friendly” isn’t the thing. Think the difference between the UI that’s inserting punch cards and deciphering punch cards as they come out vs a GUI. There might have been computations worth the hassle previously, but for many the friction wasn’t worth it. The latter gets a lot more use cases and adoption. I think this is the same kind of jump or more.
Exact timelines are hard to say, but if takover/loss of control/similar doesn’t happen first, we will see a lot of automation.
Replacing all humans works who do menial repetitive tasks like taking food order or scheduling appointments, manning tollbooths, provide assistance within stores, and more (especially with robotics – then you can do shelf stocking and food service at restaurants). Millions of jobs.
Replacing teachers and tutors, massive uphaul in education. Even if what was being taught was a dumb “if else” logic on material, having a system parse out your selections from natural language is the different in adoption vs not.
Replacing medical advice and guidance much more so Google Search already did.
This is a task for robotics, and not an easy one. I already mentioned robotics as a possible but slow-burning way this happens, and I thought you were saying not robotics, just software.
Replacing teachers and tutors, massive uphaul in education.
Ok, maybe? I buy one can get significant improvements, but mainly due to teachers being grossly understaffed compared to what is best for kids, and the possibilities of a highly accessible if fairly shitty version of mastery learning. This would take a lot of schlep / context specific building, and would still require teachers—maybe almost as many TBH.
Replacing medical advice and guidance much more so Google Search already did.
So you don’t think they’re enough capability to replace 50-100M jobs[1] in the US over the next 5-10 years? (I think this could happen from just the current generation with better scaffolding/products/diffusion, and even more so if the models continue to improve).
If you read the thread, you will see that I did not make a claim like that. I’m observing you making confident strong claims, and asking for your compelling reasons for that (and getting very little in response, which is alarming). I suppose if I were guessing, I would say no I do not expect that to happen in 5 years on the back of LLMs. It could happen in 10-20 years with robotics perhaps, e.g. several million transportation jobs replaced by self-driving vehicles that use a bit of LLMs sprinkled in, many millions of construction and manufacturing jobs made 2x higher leverage (say), many millions of retail workers made higher leverage (supervising restocking / warehouse / cleaning robots rather than doing those tasks themselves), etc. There’s a lot of “bits” jobs that I’m somewhat skeptical about them being super-duper replaced, e.g. management, finance, admin, healthcare, etc., e.g. because they are too much needing reliability or accountability or inexploitability or similar. I could definitely be mistaken about that part, but I don’t see it immediately, which is why I’m asking for some specifics. But that’s not what you’re describing, and you sound confident, but right now I think you’re just not thinking clearly about it and just getting excited because it does some cool things.
Your use case are way too general haha. They include many key things that LLMs currently don’t do. Anyway, maybe you’re not super interested in discussing whether they’re “Enough to usher in a new industrial revolution even without further progress.”, but if you were my next question would be whether the Internet would count as a new industrial revolution in your eyes. (I would say “no, but kinda / almost”, and I would say that the no LLM --> LLM transition looks like it’s kinda comparable-ish to the Internet transition.)
(I’ve been trying a new drug and my brain isn’t at 100% capacity, hence slow or limited replies right now.)
I think that’s a good question. I think the Internet doesn’t feel to me like it reorganized enough of how civilization works to quite be a revolution. In contrast to things like agriculture or steam engines where the vocation and living situation of so much of the population changed. I think LLMs, via automation, can cause an economic reorganization on the scale of agriculture/industrialization, that the Internet itself didn’t do. I’m fuzzier on where “electricity” fits.
But like, how specifically? I agree that there’s some idea around making a bunch of software significantly more beginner-friendly by giving it an LLM interface, and in some ways significantly more powerful with LLM “agents”. Is that a sufficient class of thing for what you’re referring to? I mean, do you think that 50% of people will be working on something different within 5 years, or something like that? Which 50%?
“Beginner friendly” isn’t the thing. Think the difference between the UI that’s inserting punch cards and deciphering punch cards as they come out vs a GUI. There might have been computations worth the hassle previously, but for many the friction wasn’t worth it. The latter gets a lot more use cases and adoption. I think this is the same kind of jump or more.
Exact timelines are hard to say, but if takover/loss of control/similar doesn’t happen first, we will see a lot of automation.
What are three examples of this that would be part of a new industrial revolution?
Replacing all humans works who do menial repetitive tasks like taking food order or scheduling appointments, manning tollbooths, provide assistance within stores, and more (especially with robotics – then you can do shelf stocking and food service at restaurants). Millions of jobs.
Replacing teachers and tutors, massive uphaul in education. Even if what was being taught was a dumb “if else” logic on material, having a system parse out your selections from natural language is the different in adoption vs not.
Replacing medical advice and guidance much more so Google Search already did.
That’s already being replaced without LLMs.
This is a task for robotics, and not an easy one. I already mentioned robotics as a possible but slow-burning way this happens, and I thought you were saying not robotics, just software.
Ok, maybe? I buy one can get significant improvements, but mainly due to teachers being grossly understaffed compared to what is best for kids, and the possibilities of a highly accessible if fairly shitty version of mastery learning. This would take a lot of schlep / context specific building, and would still require teachers—maybe almost as many TBH.
Yes? Seems generally fairly incremental?
So you don’t think they’re enough capability to replace 50-100M jobs[1] in the US over the next 5-10 years? (I think this could happen from just the current generation with better scaffolding/products/diffusion, and even more so if the models continue to improve).
This is measuriing in term of people’s occupations, but could instead weight it by fraction of the economy. I’m not sure how that’ll net out.
Here’s a breakdown of jobs in the US: https://www.bls.gov/emp/tables/employment-by-major-industry-sector.htm
If you read the thread, you will see that I did not make a claim like that. I’m observing you making confident strong claims, and asking for your compelling reasons for that (and getting very little in response, which is alarming). I suppose if I were guessing, I would say no I do not expect that to happen in 5 years on the back of LLMs. It could happen in 10-20 years with robotics perhaps, e.g. several million transportation jobs replaced by self-driving vehicles that use a bit of LLMs sprinkled in, many millions of construction and manufacturing jobs made 2x higher leverage (say), many millions of retail workers made higher leverage (supervising restocking / warehouse / cleaning robots rather than doing those tasks themselves), etc. There’s a lot of “bits” jobs that I’m somewhat skeptical about them being super-duper replaced, e.g. management, finance, admin, healthcare, etc., e.g. because they are too much needing reliability or accountability or inexploitability or similar. I could definitely be mistaken about that part, but I don’t see it immediately, which is why I’m asking for some specifics. But that’s not what you’re describing, and you sound confident, but right now I think you’re just not thinking clearly about it and just getting excited because it does some cool things.