The Main Types of AI
Artificial intelligence is all the rage these days. On the one hand, it’s an incredibly exciting, innovative part of the world of technology that has the potential to change the lives of human beings and make everything that we do far more effective and efficient.
However, the flipside of the coin is that artificial intelligence also has the potential to cause harm. There are several different schools of thought when it comes to this topic, some more extreme than others. Firstly, from a job security perspective. Second, the concern is that AI could possibly become a threat to humanity in more of an existential context.
Now, these fears are totally valid (to a point), but before we start panicking about robots taking over the world, we need to take the time to understand exactly what artificial intelligence is, how it works and what the different types of AI are.
AI can be separated into two different types of classifications.
The first is based on the capability of the AI in question (in order of least to the most extreme):
- Narrow AI
- General AI
- Super AI
This classification is fairly self-explanatory – it’s basically analysing the degree to which the AI is able to function.
The second is based on the functionalities of AI:
- Reactive Machines
- Limited Theory
- Theory of Mind
Let’s have a look at the different functionalities of AI.
A reactive machine is defined by its inability to store any kind of memory – thus, it cannot react according to something that’s happened in the past. Rather, reactive artificial intelligence works with only present data that it is programmed to read from the software. It reacts to the world exactly as it perceives it in the moment.
Reactive machines are given specific tasks to do, and they can perform those tasks well. But beyond what they’re given to do, they can’t do any more and they aren’t capable of learning from their experience.
Limited theory AI builds on the idea of reactive machines but takes it one step further. These machines are able to remember certain things but only for a short time – they can’t store experiences in a library of information.
Theory of Mind
Going even further, theory of mind AI isn’t something that actually exists yet, but it’s the idea that a machine would be able to understand people’s emotions and react to them in a social environment. This is the beginning of what people tend to be most afraid of.
Finally, a self-aware AI machine also doesn’t yet exist. But in theory, it’s a type of machine that would be able to have its own thoughts, feelings and make decisions without any outside interference. For instance, if it wanted to go online and get involved in Australian sports betting or start its own blog, it absolutely could.
In theory, it would have its own beliefs and because it doesn’t have to deal with human error, has the potential to be far more intelligent than humans – thus, the concern.