I am not a fan of the current LLM craze that has gripped the computer indutry, but this is the first truly interesting thing i've heard about being done with LLMs and it gives me hope (and fear) that they may yet lead to an actual AGI. We aren't there yet, but i think we have the building blocks.LLMs may just be a statistical inference engines, or whatever you want to call them, but really, is the human brain so very different from that? The real key to actual AGI may simply be giving these LLMs the chance to play and explore the world and learn and grow, the same as with human children, the big difference being that the LLM starts off with a vast body of technical knowledge.I am reminded of the "androids" from Philip K Dick's novel Do Androids Dream of Electric Sheep (and it's movie adaptation, Blade Runner). They were really just genetically engineered humans designed to have a short lifetime and be used as slaves, and in the book the thing that made them useful was they could be cre...
Related
Want to play with our latest SEAR model for unpaired RGB and thermal reconstruction? Now you can on hugging face! Bring ...
Want to play with our latest SEAR model for unpaired RGB and thermal reconstruction? Now you can on hugging face! Bring some RGB images and some thermal images and get your camera ...
Thread 👇 ❗ 🧵 #AI #slopaganda #synthetic #media #visualculture https://bsky.app/profile/bildoperationen.bsky.social/post/...
Thread 👇 ❗ 🧵 #AI #slopaganda #synthetic #media #visualculture https://bsky.app/profile/bildoperationen.bsky.social/post/3n2fvm2cxok25
💪 FT: Nvidia board not considering reinstating Jensen Huang as chief executive「 Nvidia’s board is not considering reinst...
💪 FT: Nvidia board not considering reinstating Jensen Huang as chief executive「 Nvidia’s board is not considering reinstating Jensen Huang as the company’s chief executive, accordi...