Audio version of the article
We are prone to pit emotion as an antithesis to logic and reason, but, as the embryo of artificial intelligence (AI) slowly grows, we may start seeing emotions a lot differently, probably as an amazing technology delivered by evolution for a data processing machine that has crunched the reality through billion-trillion neural networks for almost 4 billion years.
Let us understand the challenges before The Greatest Computer that exists on the planet, i.e. Life on Earth.
Life has to manifest through a form of volatile chemistry that we call biology while suffering a quality of reality we call physics that doesn’t allow perfection in anything, not even in copying.
So, the Life needs to replicate constantly to be ahead of the physics that keep changing the reality, and the most helpful ally She has found it is Data that the reality constantly produces.
By turning into a computer of data that the reality streams in all the neural networks of Life, She has, till now, managed to be one step ahead in dealing with the reality, and the most powerful technology She has developed to deal with the humongous amount of data She is processing is emotions.
Look inside your head and think about all the things you “know”.
Initially a lot of the information you may be holding (e.g. quadratic equation or laws of gravity) will look to be unconnected with emotions, but is it really true?
To understand this better, you will need to look at not the information but how you recalled the information.
The real challenge for any intelligence is not the gathering of information or processing it to find the patterns because it is of no Real Use if the same can’t be retrieved, that too in time.
Even if you “know” that a tiger kills, if your brain doesn’t deliver that wisdom in time before the tiger attacks you, you would be history. So, time is the real essence for any intelligence surviving a reality out to kill it.
Just as the natural intelligence (that has to tackle reality to survive) has its focus on the rapid retrieval, artificial intelligence too will have to deal with the same problem if it wants to compete with (and may be surpass) natural intelligence. So, it would be interesting to learn from nature about how She has solved the problem of rapid retrieval of information.
Emotions appear irrational because we don’t recognise the billion year of data crunching that has taken place in producing them. If we see them as algorithms, it is not difficult to see that they play a crucial role in helping us respond to reality but also to store data by providing rules for prioritising the data.
Emotions can be looked at as sticky notes that we use to index our documents. Just as piece of information gets stored with its own sticky note of a colour and length so we can quickly find what we need, brain dips every bit of data in a soup of emotions to tag it, and this tagging decides the priority when the same needs to be retrieved.
Human brain constantly streams in and also beholds astronomical amount of data.
It could be image of the face of your mother (or a mole on her nose (or a slightly redder spot on the mole (ad infinitum in layers we may not even know consciously)))) or the “fact” that actress X wears bra of a specific size. It could be match score of 1979 finals or who is your aunt’s second cousin’s neighbour.
The reason behind human brain beholding and yet recalling critical information in time from this jumble of information it has is that each data-piece is stored with an emotional tag.
The mother’s face is so critical that one may behold it even in old age where dementia has taken away the world from you, and the match score may remain because of the good emotions you felt that day. Quadratic equations will stay or fade depending on how you felt about knowing.
In short, every piece of data you have will have an emotional root, and the emotional strength of the connection will decide how quickly you can recall the data and also how long will it survive.
There is little doubt that, if AI will be able to deal with the reality just like natural intelligence, it will have to be able to find a way to prioritise information for timely and meaningful retrieval and hence it will have to develop something equivalent to emotions.
There is also a possibility that AI may do things differently thanks to the huge computing power it may have, but, at this point, learning from the emotions could be a great way to start not just for the machines but also for their slaves, because evolution has also produced and emotion called empathy that the slaves may need from their masters as the masters will get smarter!
This article has been published from a wire agency feed without modifications to the text. Only the headline has been changed.