They are pre-programming everyone for a military industrial complex AI takeover. Since AI can be used to help script kiddies break into php and shit like that and shitcoin loses billions of dollars. If you want to be able to run AI in the future you should at least get a copy of all the open source models.
>>545210647Won’t they just erase them from your hard drive using ai and Israeli backdoors?
>>545210647No
>>545210647Open source models are kind of s*** now, but I typically agree with your overall message. It's a weapon, it's an ideological weapon that makes finding information that much quicker doesn't actually mean doing the information or working with that information, easier for most people that don't have robots. But it's still something that you definitely shouldn't just go oh, my God, it's scary and then run away, you should engage with it and learn how to use it. It's not the AI is in the scary part. It's the people that are using AI. You know, AI is just technology that makes you smarter. You still have to fact check it. You still have to get it to question itself constantly, get it to question, what you're telling it constantly, because it'll just tell you whatever you want to hear. The real danger is just who is in control of it, which is really stupid that you have so many Democrats where no, my God, it's dangerous and then running away from it. I mean, they're the ones that are gonna be the worst off. You know?Oh yeah
>>545210839you're right compared to the major models the open source once are shit in comparison. But its better then nothing.
>>545210951Quantized Gemma 4 is pretty good
>>545211062yeah but its going to beat argon or whatever the latest shit is on the major ai companies. Not even going to come close.
>>545210647That's scary. We should focus on healing each other so that the scary part becomes less likely.
Many Claude and Gemini responses are debunked instantly using a chain of verification (COVE) prompt format. You ask it a question; it produces a list. Then you ask the same question again on each answer. Internally it checks it and you find out the initial response list is hallucinated. I realized it gave me an entire fake list for a biology question. The statement I gave it passed Claude and Gemini as true initially, but then I tried to verify and the answer was proved false. This happens because both are predictive text to generate an answer.This is terrifying if you need hard facts.There are other ways to prove it false or verify the answers.
>>545211983So for the average company or person, ot proves AI is useless & dangerous as it is literally lying all the time.If you gave a chatbot instruction to give you missile targets; it would hallucinate all of it.
>>545210647What's a good AI model to save on my spare hard drive? With no watermarking or other kikery, please.
>>545210647This makes sense.They're gambling most of the US economy on the idea people are going to pay to use AI.This wont pay off if someone's giving it away.Their best move is to kill off every other AI.
>>545210647
