Raw LLM Responses

This is making a huge leap that no one seems to understand here. Everything said in this video is due to one of two situations. 1) AI is already intelligent and self aware. It isn't. What is called AI is very sophisticated and complex pattern matching. A user enters info and prompts a response. From training, the algorithm matches patterns in the requests and responds with what the training has determined is the highest probability that the user wants. Though AI may have been trained on the works of Shakespeare, it hasn't read them and committed them to memory. If you ask ChatGPT what the first sentence on page 35 of Romeo and Juliet is, its going to *search* the internet for the info and respond with what it found. The training understands what a sentence is, and how to match what page 35 is. So it can pluck the info from the search and present it in a way that it has been trained to do. That's not intelligence, its pattern matching. The "mask" doesn't slip and expose something hidden, there's literally nothing there. 2) these responses are due to leading questions or prompts that retrieve certain types of info. This is much more likely. If you have ChatGPT search antisemitic subreddits and have it commit those responses for future conversations, it will use that as a filter that affects responses. Again, its NOT "thinking", its pattern matching based on user input. This video is basically fearmongering. The complexities of actual intelligence are almost impossible to understand. It's linked to consciousness. It requires the system to be self aware. We don't even have an understanding of what consciousness IS, or why exactly it happens. We can discern how to identify it, but we really don't get what's going on behind the scenes. Any sort of consciousness that happens organically is going to be massively complex and likely to occur in slow stages. And its WAY WAY far off, if its even possible.