non deterministic abstraction are absolutely useful, outside of software sector they have been used since the start of civilization ("a worker" is a very very non deterministic abstraction, outside from the most basic tasks)
I’m sure that in every case where there such non deterministic abstraction, it’s been always statistically or with a lot of hand waving. So with a heavy dose of expected errors.
Pro LLM users don’t want to talk about the error margins of whatever practice or product they’re putting out.
> "a worker" is a very very non deterministic abstraction, outside from the most basic tasks
Sure but we (usually) don't fool ourselves into believing otherwise. The problem with AI is that it suggests near-perfect intelligence with reproducible results. So many people are not acknowledging the fundamental limitations of LLMs, whereas we all have an intuition for the limitations of humans.
In addition to the reliability/security aspect, there's the claim to expertise and unwarranted confidence. We also don't (usually) fool ourselves into believing that the manager has gained the skills and expertise of the worker, simply because they gave abstract orders and a good result was achieved.
In other words, "I can code because I prompt an LLM" is like "I can do metalworking because I sign the paychecks of a blacksmith."
because human engineers are not reproducible, so having an accurate measure of the skills of one of them has a much lower utility return (those benchmarks are very expensive to run, and making them for humans will be at least as much expensive)
with a good enough LLM you can greatly increase the capability to build a functioning well (and other basic infrastructures) in remote and undeveloped areas without great technical knowledge, they are very good as assistants for practical tasks where you can give feedback step by step on what is happening
Well, population explosions tend to happen in hellholes and sensible declines happen in civilized society. The term happiness of course isn't something you can measure that easily, though.
Also the idea of a replacement rate of 1 equals rapidly going extinct is mathematically illiterate.
Exponentials work both ways - in growth and decay - ie let's say the population halves every generation ~ 25 years - it would take nearly 1000 years to reach near zero - as half of half is a smaller absolute reduction each time.
> it would take nearly 1000 years to reach near zero
Depending on distribution, the rate might accelerate as it becomes difficult to find viable partners. You'd also experience civilisation collapse at a much earlier stage, as we've seen how COVID affected supply chains and created a new push for regional independence and de-globalisation.
I think war caused by conflict for resources is a much more probably cause of societal collapse, and indeed extinction.
Lower numbers of people in the world reduces the chances of that.
Birth rates are much easier to reverse than climate change - it's a really odd thing to focus on.
Area's of the world where birth rate is still high tend to be in poorly developed regions where human labour is required to survive.
However, in developed countries - the idea that we need the same or more people to maintain civilisation when the means of production is being increasingly automated doesn't make sense.
As an example ~100 years ago in Europe you were looking at nearly 50% of the population being involved in food production, whereas now it's about 5% - and that includes a 50% increase in the population to feed.
Note the EU is a net food exporter - even if it does import bananas.
I don't think I can do any better than you by asking the most advanced LLM to be as pedantic as possible whilst not addressing anything actually being said.
I was responding to 2 assertions, used to justify an argument, that clearly don't hold water.
ie a replacement rate of 1 isn't some immediate existential threat, and that making arguments like 'species aren't happy' is completely specious ( sic ).
If you want a wider argument - then the reality is the birth rate is dropping because when women have the choice ( via birth control and increased socio-economic independence ) they are largely choosing to have less children. So by definition, given it's a choice - women are happier.
The second bit is speculation on motive - which is why I didn't bother the first time - but in my view, those that push an argument that there is a fertility/species crisis ( using specious arguments like the above ) - who say women aren't really happy with their choice etc - is all about subjugating women to baby making machines to satisfy the ego of people like Musk. Whether the original poster has that motivation or has just fallen for the "species crisis" ( or even worse 'race' crisis ) rubbish - I have no idea - I didn't want to assume the worse.
Hmm - childlessness ( note this is different from reduced family size ), almost doubled for women in China between 2010 and 2020, and over a similar period the female suicide rate has halved.
How do you explain that?
Sure having children is still a protective factor, but the highest risk is amongst those that wanted children but couldn't, rather than those that choose.
> Do you think average human cares about "our species is going extinct" stuff?
Yes.
> If not, why couldn't they be happy at the same time?
The same reason you can't be happy if I stick a fork in your eyeball.
You'll get back to a similar 'happiness' baseline after a while but we can all agree you would've been better off with 2 eyeballs. If you go 'well, akshually' as you are doing now, I'll stick a second fork in your remaining eyeball.
If you mind me sticking forks in your eyeballs, you should also mind society that makes people not have children.
Whatever 'well, akshually' you invent - it's deeply unconvincing.
Throw understanding images in there and it becomes less true in my experience. Sure they shit out code and search text well, but one infographic and they circle trying to rasterize it.
at the current token prices, far too expensive, but in medium/long term I think we will see a revolution in this field, the concepts are already there (and the expensive proofs of concept already work)
Those analyses tend to not consider the government interest in having AI as a weapon: now AI labs are fully part of the military sector, making them much more supported by the system
This has been my take too. I think that the revenue requirements for making the industry profitable are likely never to be met, but that the US government sees AI as a strategic necessity and will ensure that the companies that are pushing the technology forward will be paid huge sums of money if they ever become insolvent. It's not ideal, but if banks are too big to fail, I'm guessing that OpenAI and Anthropic are too. This is doubly true since the defense sector is already integrating these solutions into their systems and actively using them in wartime activities.
This is a forbidden subject here. In my opinion, the foremost subject in the entire world, but for whatever reason seldom discussed and always dismissed.
All four frontier models contracted with the Pentagon. Claude already having been involved in Operation Abduct Sitting President & Use Active Denial Systems to Make People Vomit and Shit Themselves 'because'. Institutional and corporate capture . Foreign nations now having our models trained to facilitate tailored political orientation. And the fact that these systems, for all their limitations perennially bewailed, have tremendous capacity as force multipliers for all the things civil rights activists were previously terrified about apparently just before all our problems were magically solved.
Snowden docs x³
Things that once took fusion centers, thinktanks, research, etc and thousands of manual processing hours -- now real-time, instant, with real-time 'understanding'.
I think the AI crash will be primarily comprised of collectively lamenting all the obvious things we overlooked.
reply