The Fall of AI

Home

Jun 03, 2017

A Few decades ago when humans recognized the mystery of how the mind of the most intelligent species on the planet evolves, they probably invented something which is dubbed as last human invention needed ever.

Algorithms are not only the working model of computers and machines but also of every human being. But it really isn’t algorithms that make humans give themselves the title of most intelligent species, its instinct or in abstract terms pattern. Going through the very initial stages of mankind, the highest level of human intelligence can be attributed to instinct which itself develops through patterns. While artists are well known for working on instincts, it is what helped early men invent fire, food, weapons, shelter and even language. Almost entire science is first based on instinct and then on proofs, be it the very first model of atom or the last human invention needed.

Now that computer scientists are trying hard to make machines recognise patterns through algorithms, dubbed as Artificial Intelligence, not only top players of the industry are betting high on this still far to reach technology even start-ups are evenly minting upon it. In recent years this last ever human needed invention has invented itself by leaps and bounds. From being able to correct tiniest of human (Like in this document) typing errors to travel from one place to other (self-driving cars) all by itself, machines have learned a lot and continues to do so and that even all by itself. It won’t be surprising to see these machines inventing up things needed for them to flourish even before humans could have vaguest of idea about it. Almost any Tech website on the internet has a related article that would alarm you about the rising repercussions of Artificial Intelligence, strongly backed by machine learning. Initially, the terror of A.I. has been put up only on science fiction novels and films. However, the recent paradigm shifts in the abilities of A.I. has apparently made columnists, philosophers, scientists and sci-fi cartoonists think alike. We will not only be losing jobs to automation as Historian Yuval Noah Harari makes a bracing prediction: just as mass industrialization created the working class, the AI revolution will create a new unworking class. A popular and quite practical theory by thinker, Nick Bostrom (TED Talk: What happens when our computers get smarter than we are?), could be of assigning a task to the machine which may require it to build more powerful super computers to complete the given divine task and eyeing humans as a probable intervening force might also try to annihilate the them. You might think the easy way to deal with it is to simply switch it off, but where’s the switch? Is there any switch to Internet? He asks. Now for many young curious lads getting attracted to abstract technology and talk is quite obvious. So against all predictions, theories and analysis can there be any fall of A.I.?

The reason why machines won’t be able wipe out humanity would be merely because the precision and perfection with which a machine or an algorithm works. A machine is taught everything but not to be wrong. So what will it do if one of its instinct or pattern doesn’t work? For any reason or luck, it might follow the next successful pattern according to their rank. However, it will never try something for which it can’t see a pattern. Apart from this fact that algorithms won’t make any accidents or mistakes, some stunning discoveries ranging from saccharin to X-rays are only mere accidents in course of human history. Even if A.I. surpasses human intelligence it would someday stumble upon a natural phenomenon which would require an intuitive human mind to be resurrected.