There has been much written and discussed since the early 80s about how Artificial Intelligence could one day outperform the human brain. We see it reflected in numerous Sci-Fi epics that show robots taking over the world and laying waste to the human race.

Advances in computer technology over the last 30 to 40 years have undoubtedly been remarkable and there are exciting developments getting coverage in the world press practically every day. Will AI take over the world and replace us? The truth is that the current batch of smart devices and AI developments still only account for a small fraction of the power of the human brain.

The real question we should be asking is can AI make us more intelligent?

IBM’s supercomputer DeepBlue defeats chess champion Garry Kasparov in 1997

Since 1997 when Deep Blue beat world chess champion Gary Kasparov, there has been an over emphasis on the power of these man-made neural pathways. What has been more remarkable, and less talked about, is the way many chess players since Deep Blue have used AI programs to improve their own performances. This has not only been a remarkable development for grandmasters; anyone can now easily access these programs, play against the machine, and boost their own skill level.

The Deep Blue example illustrates the stark gap between how we perceive the current state of AI and the actual reality. Let’s take a look, for instance, at how children learn in school. The best results, we know from research, come when a student has one to one tuition from an expert teacher. For most of us, this isn’t possible. You have upwards of 30 pupils in a class and the best we can do is an assembly-line kind of approach to all subjects. Simply put, it’s far too labour intensive and costly to give everyone individual tutoring.

Each one of us has a highly complex brain and each one of us is different in a million unique ways. That’s why a teacher working one to one, who can get to know a student more intimately and find a way to teach them, is the best option. The teacher can adjust their approach to make it more likely their pupil will learn the things they need to. If AI is going to be able to offer this kind of learning, it needs to possess at least some flexibility in the way teaching is delivered.

The current systems we have, mostly based on complex algorithms, mimic human cognition to various degrees. We’ve all shopped online and get recommendations from stores such as Amazon which are based on our past shopping behaviour. This type of ‘intuition’ is a far cry from a teacher working directly with an individual to develop their skill and knowledge in a subject like physics or maths, but it is a step in the right direction.

Amazon gives you recommendations based on what you browsed

So, how far we are from AI being able to take over the function of a one to one teacher?

We may, at some time in the future, have deep learning environments through AI that deliver a personalised teaching experience which suits each individual. The benefits of this cannot be over stated.

We have some exciting new developments in teaching technology as it is. Learning apps can now be downloaded quickly to our mobile phones and allow us, for instance, to speak another language in next to no time. The trouble is that these apps don’t yet offer the highly personalised approach that many pupils really need to improve their knowledge and understanding. What they do offer is the opportunity to learn and gain knowledge at our own pace, independently of signing up for a course or attending a school.

The good news is that the range of learning apps on the market right now provide an inexpensive shortcut for everyone with a smartphone or tablet to learn. A personal tutor may well be the first best thing in terms of controlling learning for an individual, but if you have the personal dedication and choose the right app you can progress your own learning experience to a much higher degree than ever before.

The future is looking increasingly exciting. If you look at the developments that have happened in AI just over the last decade, we will certainly be able to access highly personalised learning environments that make us smarter and learn quicker in the next couple of decades.