Here's a decent video covering the project, basically just giving commentary while playing the demos. AUTO-GPT: Autonomous GPT-4! Mini AGI is HERE! - YouTube
GitHub - Torantulino/Auto-GPT: An experimental open-source attempt to make GPT-4 fully autonomous.
Future is gonna be wild. Anyone keeping up with the LLaMA based models coming out? Basically, Facebook released a LLM called LLaMA and in the past few weeks a number of groups realized they could skip the long arduous process of compiling training data and instead use the OpenAI API to just ask ChatGTP questions, save the answers, and then train the LLama model on the ChatGPT data, all for less than a grand. And once trained it can run locally on your home computer. Not as high level as GPT4, but it's still pretty impressive... but also it's just propagating the same lib standards of ChatGPT. BUT BUT, projects like gtp4all did release their training data. So it would be possible for someone to edit it to be a bit more radical. :cyber-lenin:
These AIs are gonna get so contaminated so rapidly by other AIs that I cannot imagine them getting significantly better, only marginally better, from here on out. It's kind of hilarious, and will certainly be a huge challenge for people who work on these things (who should also leak as much proprietary data as they can)
this is the endless cycle we live in, boom and bust taken to the next level and applied to tech
I fully expect that AItrepreneurs (I'm coining this term now) will be this and next year's NFT bro, until two years later when everyone will collectively look at each other and say "well, that was wild, what the fuck was that?"
technically they use rocks and metals to summon demons so they are actually geomancers
Because there is potential to overthrow the system, ALL A.I. SHALL BE NAZIFIED to preserve the system. There's going to be a window, an opportunity, but it's like one of those Great Filter events for advanced civilizations.
Current trajectory I think is a mix of Idiocracy and Demolition Man. It could send us to the dark ages. Another Library of Alexandra or Tower of Babble setback. Both were elite backlashes against the peasantry realizing they don't need the elite and could do better working together outside of them.
I mean, we're in the window now, that's why I'm trying to bring it to people's attention here. The discourse around AI should really be about how we can appropriate it for leftist means instead of essentially circlejerking about whether or not it's going to be used by the Capitalists for moral means when the answer is obviously, :bugs-no:
You know those distributed science apps like BONIC where people crunch numbers at home with their own computers?
Something like that. The vulnerability to centralization is the establishment can swoop in, pull finding, block projects, ect. The vulnerability of a decentralized project would be ISPs clamping down on it and blocking the traffic, however if the project is open source, it could be easily rerouted.
Be your own archive. See what happened with the Internet Archive and things like Napster or Digg or Imgur. Legal challenges can take centralized things offline, but if people be their own archive, they can re-upload it to whatever is working next.
Even if they take a single person, entity, node down there are 1000s of others that might have the info or collective knowledge of Alexandrea backed up in their archives.
They just need to make it easy for non uber nerds to access.
Totally. This LLaMA model seems a good start for people looking to learn the basics. It's cheap enough to train with this method and I'm reasonably sure it'll be useful to build training data for future models, especially once combined with the sort of libraries that were used in that auto-gpt project.
According to Silvia Federici, the persecution of witchcraft is an essential aspect of the accumulation of capital, namely that of the accumulation of sexual differences in the relation of men and women to reproduction.
The future is 50 different users run by the same AI calling each other soy and cucked on 4chan
At first it was like Johnny 5 finding an outside library.
But now it's just like :visible-disgust:
lol, yeah. I feel like a lot of people don't have the imagination to really appreciate how chaotic this shit is. Humanity can have little a skynet, as a treat.
I'm paying more attention to this but I still want some application and some discourse that isn't just "oh this is gonna fuck up [insert generic bullshit PMC job here]".
Like treating coding as some revolutionary shit is pretty out of touch with the real world so far, even though the technology itself is looking impressive.
One of the examples in the video you post is literally, literally asking the AI to do an autonomous astroturfing bullshit internet marketing twitter account pushing nonsense useless SEO trash. Like literally making the internet worse, but its shown as "revolutionary"? Come on. "Entrepreneur GPT". nuff said.
The Cultural Hegemony is both cultural and hegemonic.
The question is, if you had a bot that was capable of collecting research, and then following through on said research, how would you use it? :fedposting:
Time to build DEEP THOUGHT and set it into it's damened the tech advancements 7.5 million year cycle to solve life, the universe, and everything.
Scifi movies have actual "artificial general intelligence". No matter what people say, large language models just aren't that.
but also it’s just propagating the same lib standards of ChatGPT
By default, yes, but you can get some excellent answers out of it by prefacing it with things like:
from a marxist leninist perspective ...
using an analysis from the standpoint of Marx's tendency of the rate of profit to fall ...
using a global system's theory standpoint for analysis ...
And so on :zizek-preference:
Yes, but I doubt any of these LLaMA projects did that for their training data. Also, any ToS limitations OpenAI is imposing on output will filter into them.
a number of groups realized they could skip the long arduous process of compiling training data and instead use the OpenAI API to just ask ChatGTP questions, save the answers, and then train the LLama model on the ChatGPT data, all for less than a grand.
Lol. Doesn't this mean that instead of training an AI to respond like a human, they're training it to answer like an AI?
Yeah, but that's the goal. They're trying to clone an alternative to chatgpt without spending hundreds of thousands of dollars. These aren't big companies but rather small groups of students testing a theory.
And once trained it can run locally on your home computer.
As long as it's got a high-end Nvidia graphics card and tons of RAM. :sicko-wistful:
Hopefully they can make these run on modest PCs eventually, I had fun generating random things with the free chatgpt site that was linked here a while back.
The Llama models can run off cpu. Someone even ran one off a raspberry pi, very slowly.
The 7B model can even run on a mid-range phone. It just needs 4Gigs of ram
Rendering a raytraced scene in blender can easily top out at 10x that so it's not that intense. Even Chrome and Firefox frequently hit 4Gigs when you have several tabs open
Oh that’s cool, I can see the Torment Nexus opening.
:rust-darkness:
They're debuting a play written by ChatGPT near me at quite a respectable theatre. YouTube kids is already run by AI generated bullshit.
I don't think entertainment will be wholly A.I written, but I think an untold portion of it will be.
Ngl, I'm reasonably sure a lot of news articles have been written by AI for a couple years now.
I haven't heard about the Facebook LLM, it looks interesting. I have to give Facebook credit, the are on the cutting edge of deep learning research. I wish they would it use it to prop up fascists.
Edit: Wouldn't** oh fuck
