The concept of what defines AI has changed over time, but at the core, there has always been the idea of building machines which are capable of thinking like humans.
After all, human beings have proven uniquely capable of interpreting the world around us and using the information we pick up to effect change. If we want to build machines to help us do this more efficiently, then it makes sense to use ourselves as a blueprint.
AI at its most basic is any technology that is designed to operate in a way that mimics how humans operate. The AI available today isn’t about perfectly replicating a human brain and putting it in a computer chip. Rather, the ‘human’ part is all about the output, or what the user interacts with directly.
Like humans, AI systems aren’t born perfect. They have to learn and adapt, and all of that is done just like how humans learn and adapt: by taking in information, or data, processing it, and storing it for future reference. It’s like when a young kid touches a hot stove. Their brain registers the pain and makes note to not do it again. AI isn’t much different.
At a basic level, artificial intelligence is the concept of machines accomplishing tasks which have historically required human intelligence. AI can be broken down into two distinct fields:
Applied AI: Machines designed to complete very specifics tasks like navigating a vehicle, trading stocks, or playing chess – as IBM’s Deep Blue demonstrated in 1996 when it defeated chess grandmaster Gerry Kasparov.
General AI: Machines designed to complete any task which would normally require human intervention. The broad nature of General AI requires machines to “learn” as they encounter new tasks or situations. This need for a learned approach is what gave rise to modern Machine Learning.
From personal assistants like Siri to movie suggestions on Netflix, artificial intelligence is rapidly becoming ubiquitous in everyday life.
Digging in deeper, AI itself is actually the largest and outermost circle in a series of four concentric circles. The next circle nested within AI is ML or Machine Learning. ML is, unsurprisingly, the learning part of AI, but ML is itself reliant on the next circle within it, or Deep Learning (DL). But it doesn’t stop there, the fourth and innermost circle, and the one that every other circle is counting on is the idea of Neural Networks.
ML is the subfield of computer science and artificial intelligence (AI) that focuses on the design of systems that can learn from and make decisions and predictions based on data. Machine learning enables computers to act and make data-driven decisions rather than being explicitly programmed to carry out a certain task.
Every part of AI is inspired by the human mind, but Neural Networks are the clearest and tangible application of that thinking. A functioning human brain is so impressive because it is made of lots of things that have simple jobs, but layers them together to make big things happen. The brain has billions of neurons that are linked together by trillions of synapses. The sheer scale of the operation makes it very difficult to replicate, but that’s exactly what scientists, mathematicians, and experts are trying to do through Neural Networks.
This Is Just Very Beginning For AI
Neural Networks are advancing, but they still have a long way to go, which affects every other piece of the AI puzzle. The advancement of Deep Learning and Machine Learning are dependent on the expansion and improvement of Neural Networks, and the pace of progress limits Artificial Intelligence. Right now, programs can appear to be human in the sense that they can learn to be more intuitive. For example, virtual AI assistants like Amy, built by x.ai, can help you book meetings and schedule appointments, Apple’s Siri can answer basic questions, and Diamond can help you manage and locate your dispersed emails and files through a centralized search engine, but even though these programs are impressive, they are not creative and are unable to make decisions beyond precisely what they are programmed to do (Brighterion).
This is why the AI that we have right now is called “Weak AI”, or Artificial Narrow Intelligence (ANI). ANI is limited to a single task and is not self-aware. It can mimic the way humans think and communicate their ideas (Siri, for example, communicates verbally and through text), but it’s limited in its ability to actually think like a human.
Where are we?
The development of AI has already come a long way, but it still has a very long way to go until we have to worry about it surpassing the abilities of the human mind. We have a lot to learn about how our own brains work before we can build something that truly mimics them, but ANI is already in use all over the place. It’s in the anti-lock brakes on our car systems, powers Google, drives spam filters, powers autopilot, and is the key to the music service Pandora and knowledge engine Wolfram Alpha. When you play scrabble against ‘the computer’, that’s ANI at work.
by Abhishek Mishra