Raw LLM Responses

I have a 1.24 million year lithium battery, simple circuits are unhackable, no memory to hack = unhackable. Ip links are not unseen, if we watch or independent detection, Sorry to say 128 or 256 bit encytion, a AI cannot knock out communications microwave backbones, a telephone company is fined, if randomly lost communications, required by law 99%, for emergency services. Bank records are not held in one place only. Funny your AI story of how AI would take over the world, the chat bot don't know how human computer infrastructure works in the real world, even if the chat bot had access to every website and video on the internet, reading the information is not the same as understanding the information or being able to hack. Any attempt to come into the real world and build a robot body takes masses amount of resources, humans are experts at tracking physical resources that are heavy and make noise and law of cannot make material from nothing. This is why the gift of knowledge, reward the AI when the drone pilot cancels target, levels of good are not being trained into the AI, a admin profile can only access. That way the AI gets a good for traget, and a great for human override. They need to expand the resolution of the rewards scale, human interaction and feedback get highest reward. The difference between a genius and a addict, is the scale of the personal rewards system. A addict needs the reward now, gets angry otherwise. These are poor AI programers, Chat GPT4 cannot beat the world stock exchange to crash it, too many regulatory bodies involved watching, and these humans are paid to watch this very trackable data. Russia's technology is so bad they need three people to be sure of real war, to turn the key, because the USA always provides a warning first, before declaration of war, thus only one person to turn key with codes, in the USA. Know your history, The USA provided full warning to Japan in WWII, provided every opportunity to deesculate the conflict, then Pearl Harbor happened.