Benn Jordan did a good video about this:

Cloudflare’s blog posts are always more interesting. They have a really good postmortem process.

The problem is that there’s no existing law that would effect this on its own. To my knowledge, no country in the world has a law on the books specifically dealing with AI models trained on CSAM. So the question, under existing laws, would turn on whether the data stored within the model itself would constitute CSAM.

CSAM laws are already overreaching in dimensions that are rife with abuse. It’s rare to have a law where owning a picture of something is highly illegal. It’s the only law I know of where somebody can just send you a picture on your phone, and suddenly, you’re breaking the law. You could be arrested and thrown in prison for even admitting that somebody else sent you the picture.

Police use CSAM as an excuse all the time to search without a warrant. Remember the duress passcode case? Police claimed he had CSAM on his phone, when everybody knew they were targeting his activist work.

And what should we really care about? The CSA. Go after the CSA. Focus on the abuse. The Epstein files are right over there.

It makes perfect sense, if your goal is to shut down OSS competition.

The core llama.cpp maintainers also work at HF and will now work for Nvidia I guess. Llama.cpp is a pretty significant part of the local LLM stack, especially since other tools like Ollama and LMStudio are just GUIs built on top.

I guess that explains why features like quantized KV caches are lagging behind. Maintainers are purposely dragging their feet.

the community may need to step up our game and get our eggs out of the big tech basket.

The community has chosen to not fight at all, which is worse. Anti-AI sentiment is at an all-time high.

Publicly. Privately, these hypocrites still whisper in ChatGPT’s ear when they get lazy enough. Or use some feature in Photoshop or some other software that they didn’t even understand was AI-driven.

We are in a war with the rich. Always have been. We will never have nice things unless you take it from them.

The way things are going at Huggingface now, I imagine we probably will have to start building the infrastructure we need to handle AI models as distributed links sooner rather than later.

Then “we” better get started. Mega corpos have fully embraced LLMs, to the point that they’ve distrusted human input too much, which is still a vital component of successful adoption. Eventually, more and more companies will wise up and figure out how to find the right balance.

Where are we at? Oh, right… we’re too busy arguing about how all AI is bad and sticking our fucking heads in the sand until the Big Bad AI Problem goes away.

We have to fix our attitudes if we have any hope of surviving this mess and not ending up as a Cyberpunk-wannabe dystopia by the time 2077 hits. Use the fucking weapons given to us!

Ditto. Kinda miss my private tracker. But, 1337x + VPN works fine for me.

You’re the one that equated “unpopularity” with “not a viable solution” with that ~1% response.

Popularity and technical viability have nothing to do with each other. It’s just that the former requires a lot of marketing to push the general public forward, way more than is necessary in my opinion.

Solar panels make up ~1% of all residential power generation. I guess that means it’s a horrible way to generate power, right? 🙄

If you watch the video, you’ll understand that the temperature transfer process to a hot water tank is incredibly slow. So, water tanks are not a very good battery.

I mean, unless you want to spend 12 hours trying to dry your clothes.

So, is this virtual VRAM stored on the other memory chip we don’t have supply for?

Not sure how that statement is right-leaning.

I partially blame the whole Oracle / OpenAI / NVIDIA cabal.

But, let’s not ignore the memory cartel’s involvement in this sham. Shitbags left themselves vulnerable to a demand crash with tiny numbers of competitors (three), tiny amounts of facilities to produce RAM, and the corporate stranglehold to keep it that way.

Not typically in the world of YouTube. That’s kind of the point.

That’s what happens after a buyout.

They are owned by venture capitalists.

Guy was a pretty good artist in his own right. He stuck to a limited sound set, in my opinion, but there was a lot of good shit from his Strange Cargo series. And yeah, I recognize Adagio for Strings in his Pieces in a Modern Style album.

I think I ended up knowing his Strange Cargo albums before Ray of Light. Frankly, he was the best part of that album. It could have been anybody singing on that one, because Madonna isn’t really a musician, just somebody with connections, with a trend-chasing mindset. Ray of Light happened for the same reason as Smashing Pumpkin’s Adore, Tori Amos’ From the Choirgirl Hotel, Garbage’s Version 2.0, David Bowie’s Earthling, U2’s Pop: all chasing that new-fangled electronica sound. At least William Orbit got good publicity from it.

Hell, I hadn’t heard from him in 25 years. Looks like he made a Pieces in a Modern Style, Vol. 2 in 2010 and started his career back up, all the way up until he died.

Yeah, even before it was released there were so many kids are still messing around with Garry’s Mod. New generations of kids playing with 20-year-old software.