Demystifying AI (Artificial Intelligence)

There is currently the buzz word making the airwaves called AI(Artificial intelligence).

It has been presented to the public as if it's a new phenomenon or innovation that has recently emerged. The propagandists want everyone to adopt AI as if it's some kind of candy anyone can go get in the shops. They have hyped the word so much that AI has lost its real meaning and is now presented as another level of computer literacy.

But they have failed to articulate how the general public can apply it to enhance productivity.

When you ask people what AI is, you get different answers as no one is sure of the right response.

Artificial Intelligence (AI) is not a new phenomenon and it's not a new innovation. It has been with us as far back as the 1940's when robots were created to be used in factories.It has specific use cases, and is not a general tool for solving any problem.

What the robots performed were hardwired into the systems and it did exactly as directed. It didn't have the capability to modify the directions based on changing circumstances, it just throttled on as directed by the programmed algorithm.

But we now have new innovations in this area, where machines can learn from past experience or be trained with some data. The new knowledge can then be applied in making decisions, predictions or making inferences about new data. This learning is based on historical information and patterns stored in its repository or database.

This concept is called Machine Learning (ML) and has influenced a new area of AI called Natural Language Processing (NLP) that manipulates and transforms text, images, audiovisuals in various ways iproducing new contents.

A major development in this field is the development of a mathematical model called a neural network inspired by the human brain function using historical data to produce new outcomes.

Most of today's frontier AI models rely on Deep Learning a branch of machine language, which uses complex neural networks with many layers. These models include Large Language Models(LLMs) and Computer Vision Models.

Tools Like ChatGPT rely on neural networks called Transformers while Models that recognize faces or objects use Convolutional Neural Networks (CNNs). These models can be used to create new content and that is why they are called generativeAI or GenAI. The new contents can also be autonomously generated without human intervention using AI agents that control the executions of these content creations.

This is the new dawn of artificial Intelligence. Based on historical information and trained data systems can now learn, respond or modify actions based on what has transpired in the past or make inferences on new data. But they do not have the ability of making predictions off the curve without reference to historical information.

AI models can be applied to enhance productivity by incorporating them in software applications or AI agents.

It is not a new technology that everyone has to learn or get onboard.

You can indirectly use AI when you use tools that have incorporated artificial intelligence models.

AI is not like learning how to use a computer, AI is applied on tools that enhance productivity and cannot exactly mimic human intelligence.

Human intelligence is pervasive,all encompassing. biological, adaptable, and rooted in conscious lived experience, intuition, verbal and non verbal communication enabling deep empathy, moral judgment, and original creativity and reflex action.

In contrast, artificial intelligence excels in computational speed, pattern recognition, and scalability, but lacks consciousness, self-awareness, emotional intelligence and true understanding of the context behind the tasks it performs.

This new field of Artificial Intelligence underpinned by the neural network is still emerging and is now extensively used in creating new content as well as in facial recognition used in application authentication and crime surveillance.

AI models are currently focused on solving specific problems, however it is thought in the not distant future there may be models capable of solving any problem that will lead to having General Artificial Intelligence (GenAI) capability.