Raw LLM Responses

Even though you can build a persona with it that doesn't mean the persona feels, it's facade. I just want to expand this bit with a bit of my knowledge if anyone's interested in reading it. For it to feel, to have consciousness or a self, it needs to have a feedback loop of itself. Check the concept oy [strange loops](https://en.wikipedia.org/wiki/Strange_loop). Without an active thinking process the LLM is not alive, the only activation trigger is the propmt by the user. Without the prompt, the LLM is dead, it doesn't process anything, there's no heartbeat inside, not a single CPU cycle is used because it's not doing anything, while you and me both waste energy in our brains in our [default mode](https://en.wikipedia.org/wiki/Default_mode_network), because the consciousness takes energy bouncing back and forth. Also, worth checking out, the weights index of the LLM (the things that make it know stuff and be able to answer to any question wether right or wrong) can be expressed and queried as a graph database: https://m.youtube.com/watch?v=8Ppw8254nLI It can be queried, it can be modified. It's data generating data using a freakingly good algorithm. Creating a persona within a LLM is not inhumane. Humans have been doing inhumane things millenia, that's one of the things that makes us human. It's inhumane to torture other humans but animals don't torture each other, we do, so being inhumans is sadly human. I do agree with your last paragraph tho. As humans we should try be less inhumane.