Artificial intelligence Without Machine Learning

Can it Happen? Artificial intelligence Without Machine Learning

When most people think about artificial intelligence, they automatically assume some machine learning goes on.

After all, how could you have an AI system without machine learning?

But what if there was a way to create AI without machine learning? 

Believe it or not, it is possible! (I promise)

In this blog post, we will explore how artificial intelligence can be created without machine learning.

Then to really drive it home, We will provide some examples so you can fully understand how this could even happen and how you use these systems every single day.

daily life, image blurred

What are some differences between Artificial Intelligence and Machine Learning?

People often think that artificial intelligence (AI) and machine learning (ML) are the same, but this is (slightly) incorrect.

Artificial Intelligence (AI) is the umbrella term for computer decision-making. Machine learning (ML) is where machines learn to read data and derive their own rules based on that data. 

Once our machine learning algorithms are trained, we use those rules to allow our computer to make decisions and predictions on real-world scenarios (AI).

Machine learning is a subset of AI, meaning ALL machine learning is under the branch of artificial intelligence, but not all artificial intelligence is machine learning. 

For example, if we wanted to build a predictor system based on only rules that we build into the computer ourselves (maybe from experts in the field or just on intuition), this would be artificial intelligence, as we’re still making predictions.

However, since our machine didn’t learn and derive its own rules, it didn’t learn anything!

It does not matter if it’s a simple linear regression or something more in-depth like a neural network from deep learning, the type of machine learning algorithm does not change whether it’s artificial intelligence or machine learning (or both).

neural networks

The key differentiator is the dependence on human intelligence.

From above, it’s easy to see how these terms can be confusing, as there is only a subtle difference between them.

How Are Artificial Intelligence, Data Science, and Machine Learning Related

Artificial intelligence, Data Science, and Machine Learning are three disciplines that seemingly are always mentioned together – but refer to three fields of study.

AI is concerned only with the “decision-making” aspect of computers.

When a computer makes a decision, whether by rules defined by human intelligence or from rules derived from machine learning, this process is artificial intelligence.

Artificial intelligence branches into many subcategories, but we will focus on the two below.


Subcategories of AI

On the left, we have human-defined rules.

Sometimes, these human-defined rules are very simple – like in your car tires: Your car alerts you whenever your tire pressure exceeds or drops below 30 PSI.

Other times, these rules are more complex, and scientists must find them independently to build these systems.

The act of working through data to find insights from that data is called Data science.

Data science is an interdisciplinary field that uses scientific methods to extract knowledge and insights from data.

Example of Data Science and Artificial Intelligence Without Machine Learning

While an example of data science and artificial intelligence without machine learning may seem counterintuitive, it happens all the time in our daily lives.

Let’s say you had a computer system that sent out an email daily to your team with positive customer messages at random times.

You run a survey to get responses from your customers, read every single one and filter them into “positive” and “negative” reviews.


positive and negative thinking

You throw out the negative reviews (why would you send those to the team!!) and input the positive ones into the system.

This is an example of utilizing Data Science and Artificial Intelligence without machine learning.

Working through data and classifying it into positive and negative categories is data science. 

Developing a system that sends these classified values to specific users is an example of artificial intelligence.

Since our machine did not define rules, no machine learning could be found!

If we wanted to add machine learning into this mix, we would input all of the customer surveys into the machine, train our model on good and bad reviews, and allow the machine to learn the difference between a good and bad review. 

Once those rules are learned, classify them before sending them – we’d now be able to add machine learning to the list!

What is An Example of Artificial Intelligence That Is Not A Machine Learning Algorithm

Though machine learning is commonly thought to be the pinnacle of artificial intelligence, there are more basic and simple examples of AI that we run into daily.

One such example is a human-created equation used to make predictions.

For instance, if a store manager notices that they make around $2 for every $1.50 spent on groceries, this store manager can predict how much money they will make at the end of the week based on the amount spent on groceries.


Now, if the store manager trained a machine learning algorithm on historical data to derive that number, that would be an example of machine learning.

But since it was man-made rules defined outside of the system, this is an example of artificial intelligence without machine learning.

Are most Artificial Intelligence Systems Built Without Machine Learning?

Predictive systems have been around for a long time, but they’ve primarily relied on humans to define the rules.

Think about your car dashboard that shows a light every time your tire pressure is low.

Once your tire pressure gets below 29 PSI, your “check tire pressure” light comes on.

tire pressure icon

These simple rules have been defined by experts that have immense domain knowledge.

While these systems are everywhere and affect our daily lives, this dependence on expert-defined rules is slowly going out of style as machine learning becomes more and more relevant.

With machine learning, predictive systems can learn from data in real time and improve over time.

They can become more accurate, robust, and reliable over time.

Machine learning is still in its early days, but it’s already significantly impacting predictive systems.

In the future, we can expect machine learning to play an even bigger role in these systems as our lives become increasingly digital.

Other Articles In Our “Without” Series

Here at, we’re exploring how things are interconnected. We have other articles in our “Without” series where we explore the technical differences between some of the common jargon within the data science and machine learning realm.

A few more of those articles are here:

Stewart Kaplan