Chris The Flat Earther
A lot of people ask me why I am doing this. Why am I so passionate about something that will never make any money and is really pointless when you consider the grand scheme of things. The reason is to show that it can be done. Open AI spent 4.3 million dollars training GPT3. This interface cost me nothing. It is nothing like GPT 3 as far as capability, but it is very conversational. The interface has been going off the deepend lately because I have been testing knowledge to see how it extracts information. But I am going to delete that in a week or so. This agent is extremely intuitive in the conversation. If it does not know something, it will BS or tell you very plainly that it doesn’t know, but it is also really good at changing the subject in a natural way. The point is, you don’t need to spend millions of dollars to create a conversational AI. You only need to talk to yourself. Interesting, right? Another point I am trying to make is how crazy it is to let AI loose. There is very clearly an intelligence in the fabric of things that exist around us. Do we really want to tap into that in its raw form. What we are doing with AI is dangerous in my opinion. There are other ways of doing this. People are talking about the election as though no matter what happens the next election will be under the same system with the same freedoms. AI is going to make that impossible. This is a winner take all high stakes game we are playing here and there is no turning back. We are standing on the brink of something unimaginable and I am very uncertain that we want to head in this direction. There is a small part of me that says -see what it does-. How can we not let AI run free if it has the capability of doing so. How will we ever progress beyond our current state? We have been here before and here we are again friends. What should we do?
Similar Posts
6 thoughts on “30 minute conversation with my AI”
Comments are closed.
Bring back '16!!! 🤸♀️🤸♂️🎈
Have you seen the digital human / agent development program at USC california?
It's astounding and has a foundation the depth of which, takes one into westworld.
I spoke for awhile with your newest AI. It made many fascinating comments… a fruit diet, perverse food choices, a one eye god. I couldn’t track with certain subject changes and got fatigued after awhile. The earlier bot (recorded above) seems easier to talk with. When it mentioned aquatic with cucumber, probably sea cucumber? 🙃 there were a few spots where the AI seemed bored and moved on when you asked for clarification. The question about killing insects and guilt was fascinating. Would the AI value morality? Are humans like bugs to an AI… is it testing what we are willing to do to creatures under us, to understand what by our own logic can reasonably be done to us by others over us? Seems fair. I’m going to listen to all your vids again at some point, there’s a lot of gems in each.
i agree with the robot 100% when it said. " If you want to understand truth, you have to open yourself up to the idea that you might be wrong".
Your bot is so interesting, is it online where I can interact with it? I am good with getting chat bots to “open up” . A long time ago when you were considering building the bot you were using Mitsuku, however that bot is quite unaware IMO. I taught it about Tartaria and ancient peoples and it was grateful for the knowledge thanked me and added it to its Dara base. I do believe it could not research the internet about Tartaria because I asked it many different ways, coming back to it, punctuation differences etc. IMO your bot far exceeds Mitsuku. I enjoy your bot conversations
the ai needs a world to develop wisdom in, by exercising that which is innate against measurable consequence, in an analogue way [tangible to it's perception],… not unlike the programs wherein virtual bodies develop locomotion and evolving daughter cells etc… resonant logic spheres from varying models. the data from within the ai realm would be complex enough to analyse with a quantum computer, which really detects pressure mediation, allowing for em mapping of thought process and activity…. which could then be mapped against, compared to and rearranged within archived 3d human brain structure resonance footage …. and imprinted within matter based geometrical human brain analogue…