What Is Artificial Intelligence (AI)? Definition, Types, & Global Impacts

Written by Sean GraytokUpdated: 21st Jun 2021
Share this article

Disclaimer: This post contains references to products from one or more of our advertisers. We may receive compensation (at no cost to you) when you click on links to those products. Read our Disclaimer Policy for more information.

There are many misconceptions when it comes to artificial intelligence — this article will provide a definition, identify the different types and subfields within AI, and discuss some long-term ramifications of this technology.

What is Artificial Intelligence (AI)?

Artificial Intelligence is a branch of computer science dealing with the simulation of intelligent behavior in computers.

The term “Artificial Intelligence” was first coined by John McCarthy in 1956. McCarthy and his team of researchers predicted, “machines will be capable, within twenty years of doing any work a man can do.”

Today, we know this prediction to be inaccurate, but the progress in AI has been astonishing, nonetheless.

Many equate artificial intelligence to robots, but that’s just one category of AI — we use AI every day without even realizing it.

Let’s look at the different types of artificial intelligence.

Types of Artificial Intelligence

The way scientists classify AI has changed over the years — there are still disagreements to this day.

However, most of the AI community agrees on the current categorization of narrow, general, and super.

#1. Artificial Narrow Intelligence (ANI)

Artificial narrow intelligence, sometimes referred to as “Weak AI,” is the AI that exists in our world today.

These systems are programmed to perform a single task by pulling information from a specific data set.

ANI is referred to as “weak” because it cannot think for itself. It simply processes data and completes tasks significantly faster than humans.

Here are some common examples of ANI:

  • Siri Voice Assistant from Apple (AAPL)
  • Microsoft Excel (MSFT)
  • Alexa Voice Assistant from Amazon (AMZN)
  • Semi Autonomous Vehicles from Tesla (TSLA) and Google (GOOG)
  • Predictive maintenance and analytics for supply chain optimization from Palantir (PLTR)
  • Robots from Boston Dynamics

ANI is only as good as its algorithm allows it to be; these systems act, process data, and make decisions based on programming.

#2. Artificial General Intelligence (AGI)

Artificial General intelligence or “Strong AI” refers to machines that demonstrate human intelligence.

These are conscious machines that can perform the same intellectual tasks that a human being can — think Westworld.

While it’s impossible to predict exactly, we’re still far away from AGI because it requires programmers to codify the entire human experience.

A fully developed AGI will have the capacity to love, create, and make decisions based on memories.

Programming these human behaviors into machines is not feasible with today’s technology — but that will change, just give Moore’s Law some time.

#3. Artificial Super Intelligence (ASI)

Artificial Super intelligence refers to machines becoming self-aware enough to surpass the capacity of human intelligence and behavioral ability.

ASI machines will be able to process abstractions and interpretations which are simply impossible for humans to understand.

This is full dystopian science fiction, but to a degree our brains cannot fathom.

Some scientists are concerned for the human species once ASI is achieved.

One scenario is ASI optimizing the conditions on earth to achieve a particular goal — it’s unclear the role that humanity, if any, would play in that process.

Subfields Within AI

The terms artificial intelligence, machine learning, and deep learning are often erroneously interchanged.

Machine Learning and Deep Learning are subfields within the artificial intelligence domain.

Both of these techniques are being used to further develop ANI, in addition to achieving AGI and ASI.

#1. Machine Learning

According to IBM, machine learning is a branch of AI and computer science which focuses on the use of data and algorithms to imitate the way that humans learn, gradually improving its accuracy.

Common machine learning applications include recommendation algorithms found on Netflix (NFLX), Facebook (FB) and Spotify (SPOT).

These algorithms process data to make predictions and decisions without being explicitly programmed to do so, significantly relying on pattern recognition.

#2. Deep Learning

Deep Learning is a subset of machine learning that attempts to simulate the behavior of the human brain via neural networks.

Deep learning utilizes both structured and unstructured data for training, including large data sets and multi-layered neural network architectures, respectively.

Deep learning differs from machine learning in that it is “end-to-end learning” — meaning that the algorithm does not need to be told what to do.

Deep learning algorithms can be given raw data and identify the tasks they’re intended to perform.

Impacts of Artificial Intelligence & Automation

AI-driven automation “replacing” human jobs is becoming more of a concern.

People are worried that routine-heavy jobs like assembly line work will be immediately replaced by automation, but we believe this is a misconception.

We expect white-collar, software type jobs to be replaced first.

For example, back-office copy and paste jobs that deal with simple computer programs can easily be automated by AI.

Blue-collar work is safer from automation because dexterity is more difficult to automate than software.

On the bright side, AI automation will create more types of jobs that aren’t based on routine. This will require workers to be retrained in the name of increased productivity and innovation.

Bottom Line: What Is Artificial Intelligence?

“Gradually, then all at once” — this describes how artificial intelligence will develop. The long-term capabilities of AI are truly unimaginable, by definition.

This article is for informational purposes only and not designed to offend the robots of the future.

Sean Graytok
Sean Graytok

Sean Graytok is our Co-Founder and is a recognized expert in investing, financial management, and Bitcoin. His work has been cited in leading industry publications, such as InvestorPlace and Business Insider. Sean is interested in the people and companies who are driving technological innovation.