Hacker Newsnew | past | comments | ask | show | jobs | submit | adrianN's commentslogin

Why do I want the model I use for coding to know Shakespeare or vice versa?

Because you communicate with it using natural language and real-world references and descriptions of what you want, you use emotion and emphasis (especially when re-prompting), you use examples and illustrative stories and common expressions. Understanding and interpreting all of that and replying in kind, to some degree, requires a large body of non-computation, cultural knowledge, or else the prompts are just meaningless words, and the replies will look like compiler output.

That sounds intuitively true, but I’m not convinced that it is actually the case. I don’t think we know enough about neural network training to say what training and how many parameters are necessary for what kind of performance on which tasks. To me it looks like we currently guess that more is better and try to throw as much compute and data at the problem as is economically feasible. There is little incentive for companies to invest into small model research since their moat is huge models that require special hardware to run.


Terse and unambiguous seem to be at odds with each other. You might want to look into Lojban and similar constructions.

Ithkuil's mad morphology allows it to pack a lot of fine detail into very short sentences.

https://ithkuil.net/03_morphology.html


I think you should elaborate a bit on that because to me it seems that EU privacy laws are actually fairly good at protecting privacy.

After a couple of years of this their expertise will be gone too and then nobody is qualified to supervise the clankers.

For my son poop is literally the funniest thing ever and he never watched YouTube.

I‘d wager that the correlation is with how exhausting the parent‘s job is. Screens are excellent for keeping children occupied, keeping them happy in healthier ways requires a lot of energy. After working a hard job, running a household and worrying about whether you run out of money before the next paycheck I can imagine that many parents just don’t have the mental resources.

> Screens are excellent for keeping children occupied, keeping them happy in healthier ways requires a lot of energy.

It could also be that the parent wants to be on their screen at the same time, or wants to be on Instagram later into the night. There will be some correlation with work, but I doubt that explains most of it.


If that was true you would see unemployed parents being best at keeping their children from the screens. It is awareness, i am pretty sure about this.

Someone who is unemployed, especially if they’re poor, doesn’t suddenly have a lot of free time and headspace. On the contrary, they just got more stressed and pay even less attention since now they have yet another urgent issue weighting on their mind.

I don't know what you think unemployment looks like, but for most people it's incredibly stressful and not a time when you can just sit on your ass and watch TV all day. The benefits, if you manage to secure them - are barely enough to get by.

From a parents perspective, I feel you are incorrect.

Almost every other parent I speak to are well aware of how detrimental screen time is to their kids, and yet often still use devices when they're too tired for much else.


This is consistent with the very old topic of television as babysitter

Any government can and does regularly throw money at industries to make them flourish. The American propaganda claims that this is less efficient than letting market forces decide which companies win.


And it turns out that the American propaganda is almost always correct.


Young people use LLMs extensively. Just ask any educator.


I don't need to ask an educator - I can just ask my kids, and they and their friends absolutely hate it.


According to Pew, almost 70% of teens report to have used LLMs/chatbots, and 30% use it daily. This is also over 6 months out of date, which feels like an eternity at the current rate things change.

https://www.pewresearch.org/internet/2025/12/09/teens-social...


You mean that after multiple years, 30% of teens haven't even ever used an LLM? Those kids probably don't even know what an LLM means since using google is now using an LLM.

But do they use it when their homework is due?


No, it's a myth - actually they scoffed and complained very loudly recently when they were picking a destination for a school trip, and the teacher suggested they would use ChatGPT "quickly" to compare their suggestions. Also anyone who thinks the kids could survive school assignments based on ChatGPT...clearly does not know who schools work these days...


Interesting anecdote, that's been almost the complete opposite of my experience in Higher Ed.

I probably know more people using AI to cheese all of their school assignments than those that are fully 'clean'.


Our experiences differ, it's no myth over here.


Having the agents write their plans into text files and iterating on those works reasonably well.


If every driver abided by traffic laws at all times we would have a lot fewer accidents.


If you’re breaking the law it’s harder to call it an accident.


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

Search: