Part 1 What's a Chatbot??

The Large Language Models of AI is an amazing tool. It is trained on the largest amount of info possible. So it will look at the entire internet for sources and references. From that it will generate new information based on statistical data that it processes and predicts the best outcome. Using probabilities and previous data, it gives the most likely answer. The biggest problem is that these probabilities can be incorrect.

I learned in this video that Large Language Models are trained from the largest amount of information gained on the entire internet. It uses math probabilities to generate new written sentences, paragraphs, stories, etc. This is all done by replicating what neurons in the brain do, which it is why it is called a Neuron Network. Generative AI creates new information gained by Predictive AI.  Predictive gets the information through patterns and previous information, then is created (images, etc) by the Generative AI.

In this video I learned that large language models are trained on the largest amount of information on the internet.  The purpose is to generate new information such as essays, poems, etc.  The possibilities are endless!  It is all based on simple math concepts, such as probability.  Using a neural network which simulates neurons in the brain, AI can produce amazing results.  

Generative AI is synthesizing data to set up predictive AI.  Generative takes in all the information to set up situations so the predictive AI can occur.  Predictive AI uses probability to predict text want to produce based on the previous information.  So in the case of the Shakespeare example, the AI used letters and words to understand Shakespeare's language and then is able to predict what words / letters would come next.  

Large Language Models are trained on the largest amount of information possible using all of the information found on the internet. Unlike other models that are trained to work on a specific task.  Large Language Models then use simple math and probability to create new content. Human tuning is still needed to insure reasonable results.

When using predictive AI, this system analyses historical trends to make a prediction. Where as generative AI can create new content that can mimic a pre existing works. For example, a new poem written just as Shakespear once wrote. 

A large language model. like ChatGPT, is a type of AI that uses probabilities to produce text such as poems, plays, essays, etc. Large language models make it possible to have human-like conversations with a chatbot. It can create large or small bosides of text, lists, lyrics, and/or answer questions you ask of it.

Predictive vs Generative AI:
Generative and Predictive AI are two different approaches to AI with different applications and purposes.

Generative AI creates new content like images and text. Generative AI is considered to be creative, like a poet or artist, and can be used to create mew information (images/text), uncover patterns in data, or develop text.
Predictive AI, on the other hand, reviews historic datato make predictions about the future.It can also validate data accuracy and completeness, and ensure data relevance and integrity. Because it can be used to make predictions or forecasts, it can help drive decisions.

What are they and hoe do they differ?
It is my initial understanding that Large Language Models use super huge data bases like the entire internet+, to predict and generate new text. It could be limited to the entire works of Shakespeare like the video exemplified or the entire internet.  
Generative AI can produce / create output based on previously trained information.
Predictive AI forecasts new outcomes based on past information.
Both use statistical probabilities but in different ways. 

1. Large Language Models (LLMs) are trained on vast amounts of language data, meaning everything available on the internet. This training allows them to generate new content, such as thoughts, essays and code. LLMs rely on basic math concepts and probability to generate new content. Additionally, these models need context provided by neural networks. These neural networks are trained extensively to predict what comes next in a sequence. Human intervention and study is available to fine tune the tool so that it works at the highest level of accuracy. 

2. The main difference between generative and predictive AI deals with the data they produce. Predictive AI doesn't create new data. Predictive data uses historical data and patterns to make predictions. Generative AI creates new content based on the patterns it has learned from the training data.