Rhaedas

Laws of physics don't exist either. What we have are our descriptions of the organized ways in which things seem to work based on our observation and experiment. But a hard coded rule doesn't exist anywhere, which is why we adjust our laws sometimes when we find new behavior through different or better measurements.

To get to your point, laws and rights are agreements within a society that things should be a certain way, and can be changed by the public or those who govern them, positively or negatively. As George Carlin said, you don't have rights, you have privileges. And those can be taken away, even the ones claimed as given by a creator. That's just core agreements, and even they aren't perfectly adhered to, especially if most people aren't paying attention.

There’s different levels. Are you looking for a filtering to avoid seeing anything potential, something on the video to indicate AI, how to tell at the start it might be AI? The first is probably the hardest to do, and if there is a way to do it you’ll either get a lot of false positives (so miss real content) or stuff will still get through. Easiest is to weed out channels as you find them. Don’t rely on YT’s little AI marker, like everything else it’s automated and will tag anything remotely suggestive of AI (but maybe isn’t).

We have friends everywhere.

It wasn't just him. But he threw gasoline on the fire. A lot.

Noted.

Most models are going to require CUDA. There are some AMD ones out there, but it's a totally different math and setup. As for the one I mentioned, it's a pretty new idea so there are only a few out there, maybe just one (Qwen based). But I did get a 31B model to work on my 12GB, I just had to move from Ollama to llama.cpp to gain the control needed to set the parameters, and fine tune what it put on the CUDA to the max it would take. I had Claude help me along the way.

It's new enough that there aren't any good abliterated/uncensored models yet.

16GB is plenty for even older model setups. Now they've got a few models designed so you load just parts of the model onto the GPU (Mixture of Experts) and use the CPU for less referenced sections, so you get both reasonable speed and a much more complex model.