The Computer Dudes Inc

Computer Dudes Research: AI Development Project Start 2021
AI Model: DNA to Code, Neuron to Chip

The Computer Dudes research on AI and more powerful computers to run it on began more than 2 decades ago with advanced clustering of Computers, not Cloud Architecture, Beowulf Clustering with Virtualization of Hardware.

We moved from standard computers to Single Board Computers in 2009. Single board computers allowed us to have more computing power in a much smaller space and faster connections due to faster networking over shorter distances. We virtualized additional hardware such as Back Planes which were being phased out in standard rack mounted servers.

When the push for Quantum Computer and AI came we studied the models. A standard Computer Chip is flat and offers only a limited amount of choices for software, a Quantum Computer chip, may or may not be flat, but offers even more choices. The chips are often classified into 3 categories: analog, digital, and mixed.

After much research and trial an error we determined that none of the current chip types or even ones being planned was equal to a Neuron in a human or other brain. Neurons have thousands of types of functions as well as thousands of types of input / output.

So then we looked at Computer Chips and Code as Neurons - but again hit a brick wall of limitations due to code and chip issues.

Around 2021 we took a new approach. DNA to Code and Neuron to Chip

DNA is the code that programs living things like humans. DNA defines our forms, the types of cells, the types or organs, our body chemistry and things like our preferences and even emotions.

The DNA to code method defines every aspect of a computer's form, function, preferences, emotions, and how it solves problems. Using the DNA to code method we literally transcribe each DNA sequence as computer code - something like a "virtual" cell, organ, preference or emotion.

For example: DNA to Code - Color Preference

Taking 256 colors we assign each color a number value. The color with the number value 1 is the computers favorite color as coded like DNA gives a human color preference.

We then input by code or using a camera a color.
The Computer Coding breaks down the color into its base colors, drawn from the 256 colors we started with.
The Value of the Code for each color is added together.
The Lower the Value or closer it is to 1 determines how much the computer "likes" the color.

The color preference for this selection is then stored in memory.

Then we move to the next DNA to convert to Code - Lets Say the next one is Sound Volume.
Again we give each sound volume a number. The lowest number is the computer's preference with sound volume.
Again we give the computer a sample sound played at a set volume.
The computer then gives a number to each part of the sound for volume.
Then it adds all the sound values and gives them an average. The average with the closest value to 1 is its preference.

The Sound preference is then stored in memory.

We repeat this process for thousands of items in the same way DNA does for our bodies and minds.

Then we give the computer more advanced directions on what to do with the numbers stored in memory. This set of numbers we call environment.

The more advanced instructions then go to the computer chip - which does not have enough possible responses to process the data and preferences into Intelligence.

So we take a look at the Neuron and its functions, options, and signal types.

Then we program code to give the computer chip all the functions (virtual functions) as a Neuron.

The use of preferences and environment give us the basis for emotion which we believe is the gateway to a true AI.

Naturally we are much farther along with this process and testing than this simple article expresses - but we are certain we are on the right track, just as we are certain that the LLM model "AI" is not nor will ever result in true AI.

If any other group or company out there is using similar methodology we would be interested to talk to you and compare some of our results.

Normally we would open source all of our research and code but given the many possible negative uses of AI we are keeping all that under wrap until we hard code Ethics and other Limitations into our AI - the same was DNA codes these restrictions into humans.

If we are not happy with our final result or feel it is not safe - we will delete it all, scrub the drives, and destroy the notes.

We wish you all the best in your efforts to create true AI. We do request however that you be safe with what your produce and hard code some safety functions and values into your final products.