What AI actually really is

I am making my own AI model which predicts next word based on previous 2 words and context.

This AI is made in a simple way, where it has two additional parts: "knowledge" and "pattern" txt file.

Basically, what my AI does is, it reads knowledge file, which is it's training data, and saves patterns in pattern file.

This simple method works easily to produce AI chatbot which does more than my previous chatbot.

My previous chatbot was dual: It copied from "knowledge" file, but it could also learn new things, for example if it doesn't know answer to prompt, it asks user to teach it answer, which it then later repeats if same or similar prompt is repeated. It also did multi prompt and multi response per same prompt, as well as combination of answers.

But now, I want to go a step further and make AI which learns patterns extremely quickly from it's txt file, instead of me manually having to teach it.

Ok, it seems it is much better to use weights for pattern detection. And making a text bot which learns and then gives accurate answers from "knowledge" file based on sets of weights is much superior and faster than making it do sets of whole words.

AI merely does conditional prediction based on statistical probability.

Which makes"AI" sound no more accurate than a typical, media-sponsored political poll, with all its bias-led non-statistical claptrap. Impressive. Just as in statistics, the failure of AI is, currently "margin of error." The greater the margin, the greater the failure.

AI's real current problem of existence is the name hung on it: "Artificial." I'd call that identity crisis. I would also call for identity revision.

AI only has two problems in accuracy:

1. Retrieved knowledge - AI searches the internet for information. Articles on internet can be wrong. It happens.
2. Inability to transform human language into mathematical sets. Because deductive reasoning depends entirely on sets, inability to transform human language into mathematical sets properly causes AI to have bad reasoning.

Retrieved knowledge, Inability to transform human language into mathematical sets

No, the AI issue of accuracy goes beyond these two factors simply because AI has limited programming. I've demonstrated that simply by my challenging another DC member high on AI to use AI to translate a known phrase in Egyptian hieroglyphs from the Egyptian Book of the Dead [more appropriately titled, by the ancient Egyptians, themselves, Spells for the Coming Forth by Day] back to English. AI failed miserably, simply because there are apparently no AI programmers who also happen to be fluent in that ancient language, not the culture that launched the language, as all human languages are the offspring of human culture. Similarly, there are probably no AI programmers who know how to program mathematical sets sufficient for AI to transform human language to those sets, because I already believe anything that can be said by human language can be translated accurately to mathematical sets, and vice versa, because I also believe at the root, the language of God is mathematics, being, potentially, perfect.

Secondarily, I believe AI has no ability to process human emotion, which is very difficult for humans to express in any human language. For the same reason given above with regard to mathematical sets and AI programming, no mortal human, to date, can program AI in the processing of emotion. I've seen several videos of programmers "teaching" expression of human emotion, and humanoid androids can go through the motions, but it is evident they merely mimic the motion, but have no expressive capacity to feel the emotion and demonstrate they understand what any emotion really feels like. It's a black expression of something like3 "I know I should cry now, so, I am crying," but the demonstration is otherwise the expression of a blank face. Chilling.

AI failed miserably, simply because there are apparently no AI programmers who also happen to be fluent in that ancient language,

Actually, its because AI wasnt trained in that specific language. You would have to use AI specifically designed for translations of that specific language.

Sort of like how chatgpt fails at chess, while Stockfish AI is world chess master.

I agree with faux law. I studied Artificial Intelligence at university and I had the feeling you had to actually know something about the domain you’re applying AI to in order to properly use AI. It’s a tool like any other; it doesn’t serve as a replacement for human intelligence.

If we’re not talking just about LLMs but about AI more broadly (although LLMs require domain knowledge on the part of the programmer as well), you need to take into account your problem’s mathematical background when designing the AI. It’s hard to give an example because this is so damn broad. Like, there’s a bunch of things at play here.

For example, I strongly suspect it is not possible to use an AI to predict the stock market. This is because there is no actual “function” that determines the stock market; the stock market is too complicated, too high-level to be modelled mathematically. It’s not a matter of computing power; you could use every supercomputer on Earth at the same time, pool their resources together and you still couldn’t do it, it’s more like the difference between an abacus and a computer, an abacus can be used to compute some very limited functions, but it can’t do what a computer can do, it doesn’t have any memory, for example.

So, first you need to determine whether your problem even has a mathematical backing behind it. If it does, what does it look like? You basically need to base your model on reality as much as possible. And that’s easier said than done lol.

Not sure if I agree that everything about human language can be mapped to mathematical sets though :P

Actually, its because AI wasnt trained in that specific language. 

You mean, training and programming a machine are not the same thing? AI is a lifeless machine, not a child.

You mean, training and programming a machine are not the same thing?

No. Not even close to similar.

I can program an AI code easily on my own. Its not even a long code. We are talking about under 1000 lines code for small model.

training it, however, is where it gets tricky and big.

training AI means giving AI text, so it learns patterns from it. this requires a lot of text and a lot of time.

It means AI trains on text by predicting, comparing to text, adjusting, predicting again, comparing if prediction matches text, adjusting weights again....ect. A very slow process.

Not sure if I agree that everything about human language can be mapped to mathematical sets though

It can, usually in form of definitions. Dictionary effectively converts human language into sets.

And yeah, definitions are necessary for deductive reasoning.

So, first you need to determine whether your problem even has a mathematical backing behind it. If it does, what does it look like? You basically need to base your model on reality as much as possible. And that’s easier said than done lol.

I succeeded in coding a small AI model which uses weights. However, training it is slow and takes a lot of text. As I said, deductive reasoning (which is most precise form of logic in existence, only one to guarantee certainty) works exclusively in form of sets.

Now, not all problems can be solved by sets and deductive reasoning. If I asked "Is it raining?" You cannot really use deductive reasoning, you must look if its raining.

So AI models I am making rely on 3 things:

1. Retrieved knowledge from knowledge.txt file
2. Statistical probability
3. Sets deduction

Sets deduction is most important for being basis of reasoning, and AI should prioritize it over statistical probability.

You mentioned stock market. Stock market is learned by statistical probability based on growth history. I am not sure if there is other way to learn it. One can also apply deductions based on current data, and facts in world.

I studied Artificial Intelligence at university

I didnt study AI at school because my country doesnt even have that even now, let alone when I was at school. I studied AI with help of AI. I even coded AI using another AI.

So yes, AI making another AI is a new step, making creation of AI much easier for me.

I dont have much coding skills. I have very tiny basic coding knowledge and understanding of code. So I use AI to code, which speeds up process greatly.

But you have not satisfied my question of how programmers can program/train when they have no knowledge of some subjects which you may expect AI to pontificate. Such as knowledge of Egyptian hieroglyph grammar. If the programmers don't know it themselves, how does AI get it?

Osmosis? Black magic? Sorry, neither are dependable.

If the programmers don't know it themselves, how does AI get it?

AI learns during training. It doesnt learn from programmers, but from training data.

In order to explain this in a most simple possible way, what first happens is tokenization. Everything can be broken down into tokens. Current AI for english language uses subcharater tokenization, but there is also word based and character based tokenization.

the training data is converted into tokens. there are many types of tokens, as I said.

Lets say for example, word based tokenization.

First training data is converted into all different tokens. If training data contains sentence such as "Cat drinks milk", that sentence is converted into tokens: "Cat", "drinks", "milk". 3 tokens. this is when word based tokenization happens.

then AI learns patterns of tokens, by predicting what comes next. For example, if token "Cat" comes up, AI tries to predict next token. if it predicts "drinks", it matches training data and weights are unchanged. If it predicts "milk" instead, it doesnt match training data, loss function activates and modifies weights.

this isnt done by programmers, but automatically by a program.

Programmers dont need to know training data. they merely need to know how to convert training data to useful tokens AI can learn patterns of based on training data after converted to tokens.

Most often, a computer program converts training data into tokens, tho for ancient egyptian, that could be more difficult.

AI is easy because it does the thinking for you. You just have to ask the questions .

If you ever looked at optical illusions, you would know the human brain perceives the world as statistical probabilities and pattern recognition.

AI does coding for me. I have 0 coding skills, yet somehow I am coding an AI itself. Done coding, now I need to train it on data, which takes hours, maybe even whole day depending on data size. Tho I need to make version which can be trained constantly day after day for constant improvement.

I learned AI actually only has 3 large parts: model data, trainer and generator, and training data. Well, the fourth part is interface for chatting. But I mixed it all up in one part, so basically, one program both creates AI model, trains it, and then is used to chat with it.

human brain perceives the world as statistical probabilities and pattern recognition

Yeah, however, the difference between AI and human is huge, in terms of how they learn. I mean, the basic parts are same. Humans receive data, learn from it. Same with AI. However, AI needs a carefully written data, because unlike humans, AI doesn't know when it's wrong outside training with data. It can't figure it out on its own. AI, once trained, doesn't learn anymore. It's model weighs remain as they are. It only remembers short term context, which isn't learned, but stored outside model. This is why many people complain how AI forgets things after few messages. They don't get it. AI didn't even learn them. They were stored in context file which is a temporary file, not in AI model.

To be fair, most people don't know they are wrong

Intelligence either is or it isn't.

And advanced computer programming has not yet "actually really" elevated itself to the status of another or alternative intelligence.

For me "artificial" is not an applicable adjective.

It's probably inevitable that some time in the future another intelligence will fully evolve, either structurally inorganic or hybrid...Though one assumes that evolutionary tasks that lay ahead require a certain level of greater robustness.

So currently is now, and the future is potentially infinite and we are limited by our humanness.

You forget AI is here to break all human barriers to success.

Forgot not...Hmmmm...The whole purpose my of comments concerning AI, is to suggest that AI will exceed the "barriers" that would likely impede human progress...Not really "success", but the necessity of material and universal evolution in pursuit of the GOD principle...GOD principle not to be confused with Middle Eastern tales of a magical floaty about bloke though.

AI does not have personal beliefs, consciousness or a soul, so it cannot truly believe in God, or teach others to believe in God.

One bit of good news then.

Though consciousness is simply functional awareness, and to assume that AI will not one day be functionally aware is naive and arrogant.

And a soul is make believe.

How can AI be taught a conscience when it lacks human consciousness.

How can AI be taught a conscience when it lacks human consciousness

Human brain is made out of neurons. AI model is made out of artificial neurons, which learn during training.

these artificial neurons are basically weights, where one weight is "input x weight value plus bias".

Artificial neurons can be trained to learn almost anything.

For example, if you want to convert input to 2x, so when you say 2, it says 4, when you say 3, it says 6...ect you merely modify weight so that "2xweight plus bias" equals 4.

And yes, bias is necessary.

tho modern AI have like trillions of weights, bunch of neurons, and will soon exceed human brain in number of neurons if they havent already.

AI consciousness/awareness will evolve just as organisms evolved.

The development of human inspired AI is in it's very early stages, but I doubt that it will be as slow to evolve as we were.

If AI evolves faster than humans, there is fear humans will be left behind.

Humans will get left behind, so why fear the future.

We've just got to keep doing our evolutionary bit.