AI, in essence, is just Maths & our love for Coincidences & Patterns

Taking an excerpt from one of my books, Scribble Vol. 1 – First Drafts, humans love coincidences:

Coincidence, a beauty
That humans love & adore
The fact that something happens
As a chance
Unexpectedly
Unknowingly
And the way things feel to be
In harmony

We admire moments where an activity in the universe makes sense to us in some format. It might not always be possible to explain why. Mathematical equations when plotted as graphs can give us designs that we’re familiar with in the real world. Flowers, stars, and other impressive shapes. Whenever that happens, we feel like the equation is "relatable". It's "telling us something".

But then there are also equations that result in asymmetric, often random-looking curves, which we conveniently ignore. Out of the many equations & graphs we come across, we tend to find meaning in the ones that, in someway, look familiar. There's an affinity towards them.

Did Mathematics conspire this to happen? Or was it our human brain trying its best to make sense out of what it perceives? Why is it that we feel some equations are “beautiful” while others that cannot be confined in our definition of beauty, are tagged as “complex”.

If trees are placed in line, it becomes garden. Random positioning of tress is “wild”.

There is a lot of Maths going on behind the scenes in AI applications. The more you witness it, the more it becomes evident that there is again a play of patterns going on here. “Sequence of words” that come out of an AI application is purely an output of a mathematical set of operations.

Just because those series of words are in a pattern that makes “sense” to humans, we find it fascinating; intelligent.

Where is the “intelligence” then? It’s an incomplete question. While Math does make the major aspect of AI, it can be said that it also plays a role in the way human brain works. Does that make it the "essence" of intelligence?

Now here’s a concern: we often see mathematical operations that, when done multiple times, move towards a conclusion, a zero value. We see this as graphs where the values move towards a minima or maxima. Since all of AI is just maths and these are just calculations happening, is it not possible that the math is moving towards an extreme end the more it gets used and operated?

There’s definitely more to it than this basic analogy about the human tendency to find meaningful patterns in maths. Let’s keep it for another article.

This is a work-in-progress article. I'll add more to it at some point, but in the meantime, if you have any ideas around this, I'd love to have a discussion. Please reach out using the Contact Me page and share this article with anyone who might find it interesting.