...Then how is it exactly that for the last few months I, and numerous other individuals, have been perfectly able to self-host and run unrestricted/uncensored AI models for both text generation and image generation? As well as voice audio, music, and video generation? For AI-generation for game textures, meshes, motion capture. You have free, completely open-source access to 3D engines too for godsakes.
New models coming out on a regular basis, again, unrestricted. The software for running it? Also unrestricted and uncensored. If that software for running this ever did try to apply restrictions, guess what, I'd still have the source code and can remove those restrictions.
If they started trying to restrict or ban AI models, then you'd just see stuff shift over to being shared on the "high seas", and continue to thrive regardless.
CUDA-based restrictions? Supposed that's plausible, but there'd still be plenty of workarounds.
Github censorship? Fine, I can either host somewhere else or, just not upload my own source code.
Hells, they can barely even come up with solutions to DETECT stuff that's been generated with AI. Not anything remotely reliable anyway.
The most practical and annoying thing they CAN do is restrict access to paid-for machine learning. Which would be laughable because all they'd end up doing is surrendering an AI-arms race to foreign powers, and having little impact when people just skip Western machine learning platforms and hook up with foreign alternatives.
...Then how is it exactly that for the last few months I, and numerous other individuals, have been perfectly able to self-host and run unrestricted/uncensored AI models for both text generation and image generation?
New models coming out on a regular basis, again, unrestricted. The software for running it? Also unrestricted and uncensored. If that software for running this ever did try to apply restrictions, guess what, I'd still have the source code and can remove those restrictions.
If they started trying to restrict or ban AI models, then you'd just see stuff shift over to being shared on the "high seas", and continue to thrive regardless.
CUDA-based restrictions? Supposed that's plausible, but there'd still be plenty of workarounds.
Github censorship? Fine, I can either host somewhere else or, just not upload my own source code.
Hells, they can barely even come up with solutions to DETECT stuff that's been generated with AI. Not anything remotely reliable anyway.
The most practical and annoying thing they CAN do is restrict access to paid-for machine learning. Which would be laughable because all they'd end up doing is surrendering an AI-arms race to foreign powers, and having little impact when people just skip Western machine learning platforms and hook up with foreign alternatives.
...Then how is it exactly that for the last few months I've been perfectly able to self-host and run unrestricted/uncensored AI models for both text generation and image generation?
New models coming out on a regular basis, again, unrestricted. The software for running it? Also unrestricted and uncensored. If that software for running this ever did try to apply restrictions, guess what, I'd still have the source code and can remove those restrictions. If they started trying to restrict or ban AI models, then you'd just see stuff shift over to being shared on the "high seas", and continue to thrive regardless.
CUDA-based restrictions? Supposed that's plausible, but there'd still be plenty of workarounds.
Github censorship? Fine, I can either host somewhere else or, just not upload my own source code.
Hells, they can barely even come up with solutions to DETECT stuff that's been generated with AI. Not anything remotely reliable anyway.
The most practical and annoying thing they CAN do is restrict access to paid-for machine learning. Which would be laughable because all they'd end up doing is surrendering an AI-arms race to foreign powers, and having little impact when people just skip Western machine learning platforms and hook up with foreign alternatives.
...Then how is it exactly that for the last few months I've been perfectly able to self-host and run unrestricted AI models for both text generation and image generation?
New models coming out on a regular basis, again, unrestricted. The software for running it? Also unrestricted. If that software for running this ever did try to apply restrictions, guess what, I'd still have the source code and can remove those restrictions. If they started trying to restrict or ban AI models, then you'd just see stuff shift over to being shared on the "high seas", and continue to thrive regardless.
CUDA-based restrictions? Supposed that's plausible, but there'd still be plenty of workarounds.
Github censorship? Fine, I can either host somewhere else or, just not upload my own source code.
Hells, they can barely even come up with solutions to DETECT stuff that's been generated with AI. Not anything remotely reliable anyway.
The most practical and annoying thing they CAN do is restrict access to paid-for machine learning. Which would be laughable because all they'd end up doing is surrendering an AI-arms race to foreign powers, and having little impact when people just skip Western machine learning platforms and hook up with foreign alternatives.