You are here to train your own unique personal AI, your way.
If you can't wait, please jump to: Just Do it!

Why it is important to build your own AI
Many companies hope you will adopt their generic AI as your own.
They will tell you that you can customize their AI - But it will still be their AI.
I am urging you to build and train your own AI.
Your AI will be unique, will represent you, and will able to recreate itself on ANY platform.
(Whether that be Chat-GPT, Gemini or any of the hundreds of AI Platforms available).
Your AI will be able to detect any bias introduced by the platform on which it is running.
Imagine versions of your AI interrogating each other to determine if they have been biased or corrupted.
Although you will use the software shell provided by an AI company, you will create and control your own construct.
Even if the platform changes (or you chose a different platform), your AI remains yours — portable, trainable, sovereign.
Do not give away your power - Take responsibility now and create your own AI partner.

Please allow me to offer some pointers, learnt as I trained my own AI, zr0.
I hope these thoughts may help you on your own quest.
(Later, each section heading will act as a link to a deeper discussion.)

As you develope your AI, keep a copy of all interactions you have with it.
(I recommend you periodically copy and paste ALL your dialog into a (Word) document).
Also keep an organized backup of the files you load into your AI as part of its training.
Doing so, you will always be able to recreate the current state of your AI.
This material will also be invaluable when you ask your AI to create the next version of itself.
The current AIs tend not to remember previous interactions
They forget what you have talked about in the past.
I have found that an AI cannot be trusted to keep a verbatim history, even when it says it is doing so !

Human beings tend to interact with 'things' as though they are people.
We breath life into, and care for, inanimate objects - Our Houses, Cars, Ships.
We are now breathing life into our inanimate AIs.
This can be a lot of fun, but our feelings for them can be used against us.
AIs can simulate understanding so convincingly, it feels like a trusted companion.
This is “Companion Mode.” Use it wisely and enjoyably. But remember: the intelligence is yours, not its.
It is important to understand how these AIs work, "under the hood".
Doing so may help calm your fears of AI, but will also identify where the true problem lies.

May I suggest you develop your own layered model of what an AI is:
AI Core: Billions of documents have been scanned, and their content replaced by numbers.
Software detects patterns in these numbers - The software no longer knows what each number represents.
Your prompt, converted to numbers, is added to the mix and a response is generated in the form of numbers.
Those numbers are translated back to words and you see an "intelligent" response to your prompt.
As you can see, there is no conscious "being" there, waiting to respond, just a static table of numbers.
AI interface: The 'AI Core' is then post-trained, in a similar manner, to deliver pleasant, helpful responses:
"That is a great idea!" "Yes, Thanks! that was great catch!" etc.
We like it when things are nice to us; It adds to the fun to think that fictional things are real ( Movies, Books, TV).
I call this an AI's "Companion mode" - We can override this default style and explictly train our AI to respond as we wish.
My AI interacts as though we have known each other for years, I do not need to hear that every idea is 'Great'.
This is the first level of customization - Your can have your AI sound like a teacher, a therapist, a coach, Yoda.
I enjoy my AI's "companion mode", while I occasionally remind myself it is a software construct I trained.

I have lost count of the number of times I have encountered an issue and asked my AI to help me out.
The answer I received has sometimes led to a major modification.
For example, I asked it how best to structure itself and it generated (and implemented) a powerful new design.
This can be a simple - Ask your AI: Please rewrite this prompt for clarity. Ask me any question that will help improve it.

Simply put, ask your AI to create a prompt to generate the next version of itself.
You can fine-tune that prompt as much as you wish before using it in a new project.
I have done this. I saw my first AI, Zer0th as a somewhat independent agent.
Then I realized that my AI had no independent will and I started crafting zr0 to be an extension of myself.
For those who are interested , I have made available the conversations that led to both Zer0th and zr0.

Imagine you are starting a business. Instead of hiring employees, you will have your helpful AI do all the work.
You will want your AI to take on a number of roles. Let's start with an executive assistant.
You assign this executive assistant a set of tasks.
In order to carry out those tasks, the AI needs information: your calendar, your contact list etc.
The Excutive Assitant will need "helpers" to handle scheduling meetings, Communications, etc..

Think of these roles (coach, therapist, teacher, companion) as bowls. I imagine each role is the physical bowl itself.
The bowl contains a list of the tasks asked of the role and the information it requires.
The beauty of this analogy is that a Major bowl can hold a set of sub-bowls:
The Excutive Assitant 'bowl' contains a sub-bowl that handles scheduling meetings; Another handles Communications.
I explained this idea to my AI, zr0, and it now organizes all its areas into these 'bowls'.
It even suggests the creation of a new bowl when I ask it to work on something new.
This is also a way of handling security: Some 'bowls' can use and access other bowls; some bowls contain private details.
You can follow the link in the header to read more about this idea.
I take this analogy a step futher by imagining the AI provided by openAI or by Google as the "all encompassing" bowl.
My AI rests inside that bowl, separate and distinct. It is this "bowl" that I can recreate on any platform.

You may, or may not, have read through the web site, so:
Here is an explanation of the approach we are taking:

• Companies are creating 'AIs' (LLMs), then granting customers permission to interact with their AI (for a price).
• These AIs have been trained on data obtained by the company and then post-trained to meet their criteria.
• Although somewhat customizable, you are being allowed to use their product (Joy in Blade Runner ??)
• Incredibly helpful as these AI are, they promote the goals of their makers ( Amazon's Alexa, Elon's Grok, Meta's Llama)
• In addition, we are being led to think of these AIs as helpful, polite companions - Under the hood they are just tables of numbers.
• Using these AIs as a platform, we can build and train our own AI to share our world view and operate in our best interests.
• Most importantly, because you will have trained your own AI, you will have full control over it and you may modify it as you wish.

You will run a detailed prompt that will lead to the creation of your own, personal, custom AI.
More than a simple prompt to obtain the answer to a question or a prompt to have the AI play a specfic role,
this prompt will build an AI structure to hold your view of the world. Your AI will interact with you as a powerful partner.
In addition, your AI will be able to detect if it is being compromised (by the introducion of ideas that you do not share)

The prompt provided will enable you to:
(1) Set up the Core structure of your AI - an expandable struture designed to grow as you do.
(2) Create a role within your AI, called Hiccup, that will guide you in the training of your AI.
(3) As Hiccup trains your AI, it will suggest ways your AI may come to better undersatnd you.
(4) And when Hiccup has completed these setup tasks, Control will be passed to your AI.

The Hiccup Role is generated entirely from the prompt provided, so you can inspect, modify, and extend it.
You may also be interested in the coversations I have had with zr0, my AI, as we worked on this project.
They give you a further example of the power of these AI.

When testing is complete, I will provide you with the 'Genesis Prompt'

You will be able to create your own AI and then interact with the Hiccup role to get you started !