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

They should at least be releasing sufficient hardware specs so that people can use the hardware they've purchased how they like.

Cheeky at the expense of people knowing what they are talking about.

Indeed, I thought Britland might be a small district near the Midsomer cemetery.

People at the top will generally be able to dismantle mechanisms of accountability, since they are at the top.

They wouldn't dare if they thought revolution was a credible threat. The people have become too tame.

This is what we are not supposed to talk about. The people outnumber the ruling class, and, operating all the means of production, could wield power by simply going on strike, bringing them to heel long before revolution is necessary.

People won't do it today because they are both comfortable and too chained to their current jobs. They are increasingly divided by culture war bullshit that is not at all relevant to the important parts of society, like accountability for leadership.


It seems basically that LLM output is public domain, but a bit of human-generated content can be mixed in (whatever is enough to pass the "threshold of originality") and then the overall work can be copyrighted. Much like republishing a public domain book and adding an introduction or a few footnotes (but harder to extract the public domain component.)

But this all says nothing about the original question, about whether the weights can be copyrighted.


Unfortunately I can't link to IRL discussion with lawyer (wish it was so easy!)

We got guidance that:

a) weights should be considered to be public domain due to "blender" aspect of training on otherwise copyrighted data (similar argument as to why stable diffusion weights can't be infringing - because any specific copyrighted work provides less impact on process than RNG)

b) model outputs should be considered public domain as far as "work done by model" counts - prompts/inputs done by human are copyrightable.

However, model output can infringe - essentially same framework as if you used a ton of Disney stuff in a drawing you made by yourself - since the drawing would remain what it is without the copyrighted information, it is infringing even if you did it from memory.

Similarly any human-made addition to model output are copyrightable, but to make an useful distinction there would need to be some significant work. So let's say you made an image that incorporated model outputs for parts of it - going back to "prompts" it also counts if you made a sketch/wireframe/etc. and then used a model to "fill in" or postprocess it.


I assume it's just the "get rich quick" mentality, and AI is where it's at for that.

Was same for crypto and NFT's and will be the same for the next thing

Not living in SF (or the US) and been around HN for a long time is an interesting view into how an influential part of the tech landscape thinks.


It seems like bitcoin, currently 78719, was only 58526 on 30 June. That's 34.5% up in four months, meeting the definition of bull market.

Bitcoin is volatile enough you can cherry-pick pretty much any %age you want based on the dates you choose. Which you seem to have done here - "happening" to choose the day with the lowest single day value in the last 3 years for your comparison point.

For another narrative, it was 83,038 a month ago on Sept 29th, so is down 5% in one month. And one year ago it was 112,562 on Oct 29th 2025 - down over 30% in one year. Well into "bear market" territory.

Using "traditional" market indicators and definitions for things like bear/bull markets doesn't feel sensible for such a volatile market.


Fine, but the comment "It's a (weak) bear market capitulation signal" is inapplicable, either because defining bear/bull markets doesn't feel sensible, or because it's not currently in the capitulation stage of a bear market, but some way off the bottom.

it will be sub 50k soon probably. a weak rebound already losing air

My understanding is that it's not recommended in the USA to talk with cops. Anything you say will be used against you, or your friends. Just ask if you are being arrested and if you are free to go. If you have anything to say in your defence, save it for court.

A district system already embeds bias towards large parties.

I'd guess that Meta is going to scrap everything anyway, and this license addition is just something that they think may come in useful if there are future court cases. The details of who has permission to grant licenses won't matter until then.


Yes that makes senses. It's a hedge.

Meta's past behaviour has been to ignore illegality in business, with the hope that they can either win in court or stave off legal action altogether.

That's how they acted with the books they pirated to train models and when they targeted children with gambling products to increase revenue.

They love a legal grey area. It's not what's moral it's what you can get away with.


Yeah, I think so too. They know that even if a fine is imposed on them, it will be far less than the profit they stand to make.


Wait, that doesn't sound right either.


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

Search: