What is Artificial Intelligence?

Author: Ian Pavlik

Do you take advice on how to run your business from Hollywood sci-fi movies? If so, then the hover car you drove to work in or the alien attack outside your office are as real as the world dominating Artificial Intelligence (AI) happening right now.

My point is that many people have a misunderstanding of what AI really is, where it stands today, and where it is headed.

You may have heard the term AI along with machine learning. They are similar, but not the same. A good high-level description from a TechEmergence article is:

“…think of AI as the broader goal of autonomous machine intelligence, and machine learning as the specific scientific methods currently in vogue for building AI. All machine learning is AI, but not all AI is machine learning.” 

For the purpose of this article, we are not going to get into the nuances of the difference and instead just talk about AI in general.

There are many forms of AI currently in play;

Self-driving cars
Traffic light coordination
Google Maps / Waze traffic routing
Aircraft autopilots
Bank fraud recognition
If you think AI is a thing for the future, you are wrong. It is all around us. You interact with it everyday and don’t often realize.

Did you do any of the following today?

Search something on Google?
Use an email anti-SPAM system?
Have Netflix suggest a new movie or series for you to watch?
Ask Spotify to suggest a new playlist for you?

If you did any of these, then you used AI.

AI’s today are very focused on doing one or a couple tasks very well. Rather then telling a computer how best to do a task, you build an AI algorithm and give it access to LOTS of data, then let it figure the best way to do the task(s). This means creating a system that can learn, rather than just follow instructions.

For your business, if you are not thinking about how you might be able to use AI in your organization in the near future, then you are already behind some of your competitors. You need to start thinking about it now.

This is where your opportunity for greater market differentiation or increased customer value lies. And why it is important for you to collect large amounts of data about your business. Without data, an AI cannot learn. You may not have an AI system right now, but when you do, the fuel to power it comes from data, data and more data.

For example, say you run 3 production lines producing different custom manufactured components to client specifications. Each run is different in volume, raw materials, finished product and labour.  What could AI determine from production data because everything is different all the time?

If you don’t collect the data, then you will not have the fuel needed for an AI to learn. If you did have data, AI might:

Predict client specific order patterns so you can proactively schedule production and Improve your customers’ experience.
Recognize productivity decreases for specific employees when controlling certain machines on particular days of the week.
Better coordinate inventory levels and allocation, which improves your cashflow.

If you are not ready to dip your toe into the AI world, the least you can do is begin to collect data. "What data?", you ask? Anything. Production up time. Throughput volumes. Resource allocation. Material inputs. Waste or QA rejections. Employee time sheets. Sales details. Customer segmentation.

Anything that is related to what you do. Track it and store it. When the time comes, the more data you have to feed AI, the better your results will be.

But how in the world am I going build an AI?” you might ask. You do not have to. For literally pennies or dollars a day, you can rent AI services from many of the big players. No need for you to spend thousands on the big computers systems you see on the sci-fi movies. Leverage the cloud and pay for only the amount of AI computing power you need.

Microsoft Azure AI 

Machine Learning on AWS

Using either Microsoft’s or Amazon’s (or other providers) systems, you can be up and running in just a few days. But you need the DATA. Without the data, the AI is useless.

“So what about the big scary AI I see in the movies?”

Yes, people are working on that, but we (humanity) are far from creating that type of sentient AI. There are real concerns around containing and controlling AI when it gets closer to being “big and scary”. But technology, computational power and our understanding of how to create brain like processing is a ways away.

For now, collect data, dip your toe and a couple bucks into learning about AI, and use Hollywood to entertain you (not educate you) on what is possible with advancing technology.

 

Published October 2018

 

go to top