For all those cheering on the copyright mafia going after Anthropic, consider that some of the groups supporting anthropic against this massive overreach of “we get to decide how you use our works” include:
Maybe this is not such a great thing?
Its an urban planning and transport issue essentially. Medium density housing (think 4-6 story blocks) allows enough people to live in an area that it becomes feasible to have trams/light rail serving that area.
You’re right that there’s orders of magnitude difference, but its the driving that’s far more! One query to a chatGPT type model uses roughly 1Wh of energy, which is about the same as is released in burning one droplet of gasoline.
No I’m a meat eater who is anti-car! I’m more getting at how people have latched on to the energy use of AI models without realising the huge energy usage that goes into their daily lives.
Yeah, I too hate those hypcrites who complain about the massive environmental impact of AI, then drive a 10 mile round trip to buy a burger made from a cow raised on soy.
Just FYI, this use of republic is not recognised in political science and as far as I’ve seen is only used by americans justifying why their system is undemocratic. Republic just comes from “res Publica” (public affair) and means the head of state is not a monarch but a member of the public. There are very democratic republics like Finland and there are very undemocratic republics like the PRC. The way you describe a republic would apply to countries like the UK or Sweden, which are constitutional monarchies, not republics.
Representative democracy is a better term for what you are talking about, where the population elects representatives who are able to advocate for them and take the time to become subject matter experts on running the country (idealy).
The most annoying thing is there is a trivial to implement way to close to halve these stupidly inflated sizes: make the highest resolution of textures a free DLC that you optionally install
Yes, i find it difficult to believe that they mess up a dozen line algo that is in their training set in a prominant place with no complicating factors. Despite what a lot of people here think, LLMs do have value for coding. Even if the companies selling them make ridiculous claims about what they can do.
I find that very difficult to believe. If for no other reason that there is an implementation in the wiki page for Levenshtein distance (and wiki is known to be very prominant in the training sets used for foundational models), and that trying it just now and it gave a perfectly functional implementation.
Yep, and when you click a button that liteally says “make this discoverable on search engines” which is off by defualt, its the later.
deleted by creator
It would be nice to see a price/GWh of this (along with running costs, it says they save 1 Million per GWh, how much were the running costs before!?), but any improvement in battery tech is definitely a good thing.
It wont do anything of the sort. Even if you accept the premise that somehow artists are being exploited from learning from their previous works, all that will happen is the AI companies will shift out of America to a juristiction that doesnt value extracting rents from IP above all else.