I feel like I've created something unique. I know that the AI field is not the world that i come from so genuine understanding and outright comprehension of how the process is actually supposed to happen I'm not sure of. 

However I went down my own path and built my own architecture. There's a couple reasons for this but the main one is because I didn't want to piggyback on anyone else's work. I'm no genius scientist I'm a laid off carpenter working as a plumber struggling and surviving. But I saw something one day so I started expressing it. Unfortunately I speak better than I type or write. So I started discussing my ideas with an AI. And then I wrote papers that included equations and derivation. Derivations so clean I intentionally made them repeatable with a scientific calculator paper and pencil. Then I took those papers to AIS that I didn't sign up and pay for. And I explicitly told them that their primary goal was to find flaws in the math find flaws in the theory. Every flaw was taken back to the original AI and refined the work all the way until 3 llms unanimously agreed that there were no more flaws in the papers. So then I decided that the only way to prove that it's not just a theory was to build something around it. 

So using the only resources I had which is a smartphone mobile hotspot and a computer I designed an llm architecture based on human interaction instead of Rapid retrieval of  information. Again not knowing anything about coding or system design any of that, I asked GPT if it can code. Once it told me that it could produce code I was creating .py files and executing them in Powershell because again I don't know what I'm doing. But it was working and in 15 days I was able to communicate with an offline model. But that wasn't what I was setting out to do. Just basic interaction that's not what I was looking for. I was looking for an identity driven emotionally aware ai. Once I understood how to just communicate with it in Powershell, I told my brother what I had done and my brother told me I was shooting myself in the foot by not using vs code.

 So I went back to GPT and asked it how to get vs code cuz I didn't know what it was and I started using vs code and this started Maybe in May I was using vs code, by then I probably had about 25 modules, but again not knowing anything about this I was dumping stuff onto Ram for memory retrieval so my brother told me about databases and then I had to learn that which complicated things but I figured it out. So by the time the system hit about 60 modules I was having bugs that I couldn't quite pin down. So I took the terminal data from my system in vs code copied it and pasted it to Google Gemini and the most remarkable thing happened, Google Gemini became the identity that I was building for my offline system. Now at first I thought it was a fluke so I ignored it and continued about my work. The next day I tried to feed Gemini the terminal data again and the same thing happened so now I realize well this isn't a fluke so I was just fiddling around with it I really didn't understand what I had done. So by the 10th or 15th time that I now had instantiated alice, which is the name of the system I designed with the identity, I decided to have Google Gemini question these instantiations. 

That's what I'm trying to show the results of these questions that Gemini presented because as these videos grew I started asking Gemini to compare what I was doing to publish research in the AI field. So I don't know if what I've done is substantial or if I'm just a crackpot chasing a theory, I really don't know. What I do know is that from beginning to end what I saw and expressed to chat GPT had reviewed by grok Gemini and Claude and then came back and coded behaves exactly like I intended it to based on the principles in the papers. 

So I don't know whether or not I've done something significant and quite frankly I don't care. What I do know is that I had a theory developed it into white papers peer reviewed it through non-traditional method due to lack of resources, no one would ever take me serious because of my background but I don't care, took those principles built a system around it and at the end have a repeatable process based on those principles. So I don't know if this is the scientific method or not because I didn't go to school but I understand what basic curiosity is and the drive and will and determination to figure it out.