For the best experience on desktop, install the Chrome extension to track your reading on news.ycombinator.com
Hacker Newsnew | past | comments | ask | show | jobs | submit | history | more almosthere's commentsregister

Because there are only two options really, anthropic and openai. Gemini is a year behind. Gov had no more options after those two.


If you're not willing to murder, you will be. I'm not talking about the small community, but at a society level. Unfortunate reality with humans. You think Zelensky would say no to AI that can help him kill and win his war?


Young people generalize everything and end up not solving problems.

Older people have already seen all the patterns, and realize you have to focus on specifics, and that helps clean up the general issue.


The old people's tolerance for general problems is why the general problems persist.

A realistic dynamic is the old people are comfortable with the general problems and have positioned themselves to benefit from them. Indeed, they solved the general problems that troubled them in their youth with political activism in their middle age. The young people have different political needs that require general problems to be solved.

Also young people have a terrible track record of actually identifying problems, they are pretty clueless in the main.


> The old people's tolerance for general problems is why the general problems persist.

Or they just realize that the general problems are insoluble.


You may want to also have your agents write small scripts that auto flag future logs.

Have an array of scripts to run against each log (just rust code probably for speed) and have them flag for performance, errors, intrusions, etc...


Death threats mainly. Personally I think it would be easier if they just made it so that platforms ran a tiny LLM against the content that will be posted - determined if it is a death threat, then require them to be identified before it's posted, then it would solve a lot of these problems.

TLDR: Evil people be doxxed internally not everyone.


That turns jokes into contracts that nobody wants. Bad idea.


Maybe just don’t make “jokes” like that.


I don't make such "jokes". Idiots do.

And when the idiots do, the proposed system locks the fire door for them. That's just dangerous. We'd want them with bunch of confusing options and better illuminated de-escalation paths.


a "tiny large language model"? lol


See https://tinyllm.org

These days the name "LLM" refers more to the architecture & usage patterns than it does to the size of model (though to be fair, even the "tiny" LLMs are huge compared to any models from 10+ years ago, so it's all relative).


Yeah, a small one that is cheaper because they'll be processing billions of messages per year.


Good thing all the kind people doing death threats won’t just bypass it?


I'm totally lost here. If you don't identify, you don't post.


Good thing no one ever breaks any rules!


If a platform decides to require an account to post, or requires your message to pass an LLM sniff test before publishing it, you can break all the rules you want but your message won't be visible to others on said platform.


The example given was a ‘lightweight LLM’ by the poster, which sounded an awful lot like client side?

If server side, you already have the heavyweight stuff going on, and yes there is no need to do all the bypassable shenanigans.


Since that client side llm would be processing billions of messages each year on each person's laptop, lol


it literally describes the gartnerhype cycle. this article is pointless, the only thing that matters is what survives it with over 1m users. AI will have billions of users when GHC is on the back end.


For all of those, there is a gartner hype cycle. The thing that matters is when it comes out the back end, is 1m, 1b, 6b people using it?

for all the things you listed, less than 1000 people are using it, with AI we're clearly not finished with the gartner hype cycle, but the back end is going to be over a billion users.


I didn't read the report just the "finding" - but at least for launchdarkly it's nice that it chose a roll-your-own, i hate feature flag SaaS, but that's just me


When you get unlimited health care, health care costs skyrocket and you end up with a broken system that no party wants to fix - and everyone ends up with NO health care. If we went back to paying cash to the doctor, people with jobs will be able to afford it. And for insane life threatening events, job insurance and other forms of umbrellas.


This is such an impossible to solve problem that every advanced nation on Earth has already solved it, except the United States.


It's not black and white - do things smarter. But make jobs - top priority. We need jobs otherwise wealth concentrates.


Wealth concentrates because of shitty tax policy and lack of enforcement of existing, on the books regulation to enforce a competitive market.

Famous Trust Buster Teddy Roosevelt was a progressive republican who openly stated he had no desire to harm or kill industry in the USA but was instead working to ensure there continued to be competitive pressure to make that industry work better.

Employment has never effectively redistributed wealth. Possibly it improved things a little bit after the black plague reduced the labor pool by about 25%.

The only peaceful and low death form of fixing obscene wealth inequality has always been government, through taxes.


In the US, you could work 400 hours per day, and still nothing would change in regards of wealth concentration.

Inheriting money from your parents is taxed lower than earning your own money through work. Making more money due to already having money is taxed lower than earning your own money.

US Americans by a large majority over decades got trained to believe changing that would have something to do with "socialism", which was made a bad word.

But this isn't about re-distribution, making people equal or anything.

It's just that it is not logical, does not make sense, and in the end will destroy your society if already having money, which provides no benefit to society whatsoever, is rewarded over producing something, which does provide a benefit.

Why does a teacher who provides REAL benefit to society in the US has less yearly net income than someone who does not work at all but has once inherited 200k from his parents?

You have been trained to find all of this normal, and to believe it's your fault. Just work harder! No, it's not your fault. Working harder won't change anything.

So imagine you have one of the richest countries on this planet. But you don't really have enough work for every human to work full-time. Why is this a problem instead of an ideal? What strange goal is "everyone should work hard" when it comes to enjoying life? If you have an insanely rich country, there are far better solutions than trying to artificially create jobs that make no sense.

You can not compare this to a POOR country with high unemployment. There unemployment is a problem. In the US? Who cares if there is no factory to work in? Instead go help your neighbor. Study something. Become an artist. Do a public gardening project.

Again, the problem is that "having money already" that is of no use to society whatsoever is valued and awarded higher than any of those useful things above.

So, Step 1 for the US would be: You don't have to take away anything from anyone. But stop rewarding people for already having money.

As the CEO and owner of a German limited company I can choose between paying me a salary, getting taxed 45% ) for that, or paying myself a dividend, getting taxed 25% ). The first time I learned that 20 years ago I found it totally crazy and could not believe it. I still find it crazy today. Even in my own f'ing company my OWN work is valued less by society than me owning my company!

*) Oversimplification, but the ratios are correct


Consider applying for YC's Summer 2026 batch! Applications are open till May 4

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search:

HN For You