Live data from Hacker News

Ask HN: Are we as society going to let LLM companies take all the values?

news.ycombinator.com

11–20 of 20 posts

Re: Ask HN: Are we as society going to let LLM companies take all the values?

#11
About 200 years ago, someone figured out a commercially viably way to make yarn into cloth using a machine, instead of by hand, and it created a significant immediate disruption, but loom-generated clothing quickly became a commodity that made life better for everyone.

That wasn't the last time that a high-labor field was automated, much to the chagrin of current and potential future workers in the field, only for the once-expensive output to become a commodity that significantly raised the quality of life for everyone. Despite the Luddites fears of an economic collapse, it always grows.

The only reason we have desk jobs is because we've automated so much that only a small portion of us have to work on things we need to survive. Most of the tech industry is really an entertainment industry that isn't at all economically necessary. As our culture shifts around the endless entertainment possibilities available, we change what entertainment we want to throw money at, and people do more of that, and less of what they used to do, with minimal impact to the rest of us.

As long as there's enough people and machines making our food and clothing and housing, we're all going to be fine.

Granted, there are plenty of efforts to stop people and machines from growing food or building housing, but that's an entirely different problem.

Re: Ask HN: Are we as society going to let LLM companies take all the values?

#13
post #12

It’s just a tool dude/tte. You’re caught in the hype, and it’s all around you in your internet bubble. Tune out, listen to some music, watch a movie, ring your Mum/Mom and ask how her day was.

A tool with a monthly subscription isn't a tool, it's a service.

Re: Ask HN: Are we as society going to let LLM companies take all the values?

#14
post #8

> All the values (measured by money) created in this world is sucked by the LLM owners/producers Intellectual property was always a farce. It really is as simple as that.

In the public sphere yes. A long term effect of this will be what ends up in the public sphere won't be the same as it once was.

Re: Ask HN: Are we as society going to let LLM companies take all the values?

#15
I'm with you but I see a bigger problem. I'm not sure if you noticed but there are threads of low desire within our society. Young people, in particular young men, increasingly seem to be checking out of dating and higher education, both examples being paths towards enriching oneself. As these people age, lose mental plasticity, and become entrenched in their ways, how likely are we to be able to convince them to begin valuing those things again?

I'm less concerned about the impact of AI specifically and more about concerned about people simply losing the will for one another or anything at all. None of this stuff in the real world will matter too much if people are getting their basic needs and desires met in a virtual one.

Re: Ask HN: Are we as society going to let LLM companies take all the values?

#16
> People stop creating music, stop writing blogs, stop doing experiments or any other cool stuffs and share it on the internet because LLM companies can just pirate the shit out of it without paying anything back.

This isn’t necessarily true. I still write my own blog by hand and write code by hand. Sure it gets fed into the LLM meat grinder but I think a lot of people are going to fight against the LLM slop and consume real human content

Re: Ask HN: Are we as society going to let LLM companies take all the values?

#17
post #13
post #12

It’s just a tool dude/tte. You’re caught in the hype, and it’s all around you in your internet bubble. Tune out, listen to some music, watch a movie, ring your Mum/Mom and ask how her day was.

A tool with a monthly subscription isn't a tool, it's a service.

Don’t get confused on the semantics.

Service / tool, it’s just a thing you utilise.

Re: Ask HN: Are we as society going to let LLM companies take all the values?

#18
I'm mostly with you. The challenge in societal and economic change this time around will come much faster than the last time(s). If we (= our politicans) don't get a grip on the downsides of technological development, from reigning in social media and ads to distributing wealth and opportunity more equally I fear we will see a lot of violence and high chances we end up in a world like Bladerunner (1982). It's not so much about AI, but about the acceleration and polarisation of the downstream effects of technology that carry the seed of revolution and even violent revolution. That said, we can still start to turn the ship, but I don't see it. Global politics has all the hallmarks of preserving the status quo, preferably with personal enrichment, consequences be damned. So, into politics you go, fight for change and hope that you'll make it just in enough time to ward off the worst.

Anecdotal side remark. Yesterday, while cooking, I saw a NYT interview on YT with Tucker Carlson and he - of all the people - came to the same conclusion, albeit from different premises which did nothing to soothe my worries. If people like him see economical divide as a possible precursor for a revolution things have taken a turn for the worse.

Re: Ask HN: Are we as society going to let LLM companies take all the values?

#19
Fellow LLM doomer here. Don’t forget how many people uncritically accept LLM responses as true, unknowingly read LLM-generated articles, or listen a Youtubers reading LLM scripts as if it’s real thought/opinion. I saw an obviously LLM-generated Steam review for some reason today. It will be 100x worse when even people hyper aware of it can no longer tell what’s real.

Whoever controls the model weights will just get to control information, data, and opinions of the masses going forward I guess. This is going to be my generation’s Fox News one day.

Re: Ask HN: Are we as society going to let LLM companies take all the values?

#20
you might have the same issue as I had until I noticed the obvious:

1) people assign value

2) we've grown to 8+ billion and we're not gonna stop

3) everything is made up, every rule of value and engagement beyond pick, bite, chew, shit and kill when hungry or necessary is complete nonsense and only works because we, across generations, collectively decided mostly on "ok, I can make this work for meself"

4) almost all markets grow either way, meaning whether AI corp employees give AI users more time/attention on the marketing stage as people who don't use AI or not

5) people won't stop doing anything, worst case scenario ( "objectively" ): they take pills and then do that or some other thing, despite theft or any lack whatsoever

The thing is: it doesn't matter if something is just a prompt away. prompting and so on equals time equals money. I made a website for someone via claude designer ( opus, then fable ) ... now I want to hire a web dev to iron some kinks out and polish some others. I can do it myself and I can prompt claude again, but if I had a 1000 bucks, I'd rather pay someone, EVEN IF THAT PERSON WASNT A PRO BUT ONLY TOOK THE TIME BROWSING THE WEB WITH AI TO GATHER DESIGN INSPIRATION WHICH I CAN CHECK IN THEIR PORTFOLIO.

Some peeps learned code for fun, others for profit, others because they hated MS Office. Most people didn't learn to code. Most people won't prompt an AI to build them a tool they want because the supply just isn't the right match.

The logical conclusion of LLMs is this: literally nothing changes. The economy will always have space & money for people who provide some sort of productivity. And "consumption" has been an obvious productivity metric forever, but you have to do it at least somewhat publicly. Then the ads industry can help you make some bucks, while AI helps you with SEO and stuff, ... or you hire someone or use some 10 dollar per month saas.

If human input stops, LLM training stops. "Agentic World models"? One bug and they are all just tin cans. "Why? They can self-debug!" They can, but without humans they reach dead-ends, "task done", "nothing to do here" just like MOST employees in any corp ( SO is down and the 69x dev is on vacation ).

These researchers you talk about work. Enough of them make music.

Rich peeps bought land forever. Hell, my neighbor is poor and owns some land he bought for 5k or so. Lots of wood. He recently came around and now likes AI, after years of hating without even trying to fix his time-and-motivation-dependent ( learning ) issues with AI even once.

I'm modestly certain AI will drive enough people back out there into real life. Things will balance out.

AI has created no new issues. Most young people's problems are mid-to long term results of cheap liquor, bad food, downers and blockers ( experiments, parties, hopes, sabotage, side effects ) and the psycho-cultural effects of too many lies on TV, too much fraud in the economy and politics, and too many stimulant-boosted kids in school and university. Pills can solve their issues but you'd need to check and monitor their biology and neurology very closely and we don't have the "man"power for that, yet.

#volatileGasPricesWhatWhy #dontTryToExplain

Post reply on HN