Skip to main content

Futurist: On the Road to Machine Intelligence

I am a long-time Futurist, and technologist. In my career, I have spanned the birth of personal computers, to the rise of Cloud Computing.

Sometimes it is important to look past the presented image.

Sometimes it is important to look past the presented image.

Automation, Machine Learning and Artificial (Machine) Intelligence are not the same thing

As a futurist, I often look at technologies on the horizon. Last week I released the 2023 technologies that will be big. Or technologies that will be bigger than they are now. But looking at the horizon, I often reflect on the current state. One of the things that I find incredibly interesting right now is that virtually every product in the technology space advocates that it is artificial intelligence. Most of those are simply the pattern matching of machine learning. Machine learning is part of Artificial Intelligence but, on its own, is not intelligence. As these things continue and they are continuing, the market will continue to grow. But it is very important to differentiate between the three things available today: automation, machine learning, and machine intelligence.

Machine learning and the newer deep learning are extensions of the concept of pattern matching. Data is collected and made available. The computing asset can then review the data and see what patterns might work on that data set. The pattern is applied when the machine finds the pattern that bests maps to the data. Applying that pattern to information is really what machine learning is all about. There are several machine learning patterns, and you can use the pattern for the data you are analyzing. What machine learning does he quickly apply the rules of that pattern to the information you gathered? If you think about many things on the market today, they call themselves artificial intelligence-driven; in reality, they are simply utilizing the pattern matching of machine learning.

Automation is a simplification of a task. Why is it simplification? Take common tasks that must be done and automate them, making them easy to repeat. One of the most hated automation applications today is the 800 support bass that many companies use. Those five or six questions you answer before you speak to a human being are in the automation system. Structurally many automation systems are called machine intelligence which is simply automation. Many tasks in creating a self-driving car are automation, not intelligence. That's why today, most hands-free driving systems come to a point where they prompt humans to take over. It is because there is no true intelligence running the car today.

The reality of artificial or machine intelligence, particularly today, is that any system using machine learning and automation to drive a new answer would be machine intelligence. Let's stop briefly and define the term driving a new system or answer. With automation, you are making a complex task easily repeatable. It is the process of doing the same thing over and over. With Machine Learning, you are applying patterns to data. Machine intelligence is the ability to create a new answer to a problem that didn't exist before. For example, let's take AI-driven video cameras. Most "AI" cameras today use machine learning concepts to find a human face. The original versions of those cameras used voices as a model to direct the camera. Simply the auditory clue of a human speaking means you turn the camera to the speaking human. A camera capable of delivering a machine intelligence system would have the following capabilities. One of them is if you are unable to or unwilling to pay full attention to online meetings, Based on context clues and patterns, take the automation to the next level and apply it differently. Instead of focusing on you looking at something else, replace your live image with a previously stored image of you. That would be a machine intelligence version of the technology. You go beyond facial recognition to recognize what activities are holding your attention. I know in the online meeting world, it's called multitasking. But you should be able to multitask with your camera, knowing you're multitasking at applying a previously stored image of you staring intently at the screen. That would be machine intelligence driven.

The reality of our modern world is advertising. AI or Artificial intelligence is a great buzzword. Since the definitions can be smudged, artificial intelligence or machine intelligence is a buzzword used in advertising. Eventually, it will truly have meaning. When it has a meaning and builds systems that provide value to the people using them, and the value isn't a simple pattern, not sure automation of a series of repeatable tasks, then you will have the rise of true machine or artificial intelligence.

This article is accurate and true to the best of the author’s knowledge. Content is for informational or entertainment purposes only and does not substitute for personal counsel or professional advice in business, financial, legal, or technical matters.

© 2022 DocAndersen