Raw LLM Responses

Disinformazion, speculation given as reality, uncertain and not probable future as certain, this video is just a movie for you to watch, has nothing to do with reality. Llms are just word generators, very complex, very good but just that. If you train it with the entire X database you will maybe see that in some occasions your model will threaten you, or communicate like it is thinking but its not, its just spamming words. I give you an example, if you say to chatgpt something like "please past your imagine file here" it could "think" that it is inside of a power point and it will continue giving instruction on the work. It is just a word generator, if you say "my dog is" It will say "Black" "loyal" so when you say "chatgpt are you a Monster?" or "are you here to kill humanity?" Whats the most probable sentence after those? You tell me why your Instagram is full of videos of llms saying these bullshit, dont believe its like this. I dont say there is not space for a distopian future with dangerous ai controlling us but you have to free your knowledge from this bullshit and undertsand how easy things are first, they are just word spammers or pixel spammers or frequency spammers in case of video or audio. Think of this video as disinformation source that uses uncontexted videos and infos to build some facts that do not exists. As an example, when they say its "non human" or "alien" intelligence, it is not really scary, it is a way to speak to make YOU undertsand that machine learning chat bot you text and talk with, do NOT think like you. It is certainly not to say that we created some alien, scary or distopic intelligence, just to say that how a llm think it is different from what you normslly would think, in fact machines dont think as we think. Instead in this video the author took those clips without context as Openai's CEO is terrified of having created a Monster but the reality is that he is NOT. They are excited for the power and usefullness of the tools they are working to build and yes there is chance and people that say that future form of ai could be dangeours but for now the dangers are the same you have with internet days, no privacy of your informations, people will tend to be lazier and more stupid because they will work less for a X result with ai doing the work for them and safety because if you ask gpt to hire someone it could accept thief and killers as babysitters because its an AI and its stupid and it make mistakes. We are not scared of ai, not in the sense of this video and yes for NOW but the future you are looking for to be terrified is far and non probable.