Today’s intelligence test: which is the real AI?

 

I’m an expert on AI. A few years ago, I set up a business called AI Management Consulting. I aspired to use the amazing capabilities of AI to solve the world’s problems. The business lies dormant, however, while I ponder whether the world was ready for me.

A friend, a former professor from one of the top universities in the world, congratulated me on my appearance in the AI space. He himself is literally a world-class AI expert (he has 6 degrees from some of the best universities in the world), and owns and runs a cool global digital business.

I was very flattered of course, though I had to admit to him that the AI in the business name stands for Adzhar Ibrahim, not Artificial Intelligence as you might think, and as he did.

Artificial Intelligence is today on everybody’s mind, partly because natural intelligence isn’t. One day, it’s possible natural intelligence, along with the natural world, will disappear, to be replaced with Artificial Intelligence and cockroaches.

-ADVERTISEMENT-

Ads by 

Having established my credentials on AI, I feel I need to address the major existential issues facing us because of advances in AI. It’s only fair as some may associate me with this blessing (or curse) that’s been unleashed upon the human race.

For most people, AI entered their consciousness and vocabulary late last year when something called ChatGPT was launched. Overnight, everybody became expert in it and bored everybody else to tears with their expertise.

How could something with a name as silly as ChatGPT be of any consequence? A fair enough question. However, we’ve learnt to our cost that silly names for revolutionary tech can hide their immense disruptive power.

The AI menagerie

Who would’ve thought some silly stuff called Google, or Facebook, or TikTok, or even WiFi, would have such impact on our lives? It wasn’t that long ago that big tech had proper names such as “photocopiers” or “International Business Machines”.

What’s in a name indeed.

ChatGPT is part of a zoo of new and not-so-new products with names like Bard and Bing. I’m waiting for a new product called Boom and for the whole menagerie to merge and consolidate so we end up with Bard the Bing and Bard the Boom.

Thank you, thank you, thank you. You’re a great audience! You can sit down now!

AI (the computer thingy, not me) has become so advanced because of major technology leaps in computing software and hardware. On the software side, people discovered that programming computers to be even smarter can’t quite take you to the next level.

The secret then became a matter of how to make computers think like humans. Some smart aleck somewhere developed what’s called the neural network, giving rise to a bewildering set of new possibilities.

Can machines think?

The AI thingy in ChatGPT and other programmes run on something called Large Language Models, which aren’t overweight students from the English Lit faculty jumping on stage. Rather it’s something that treats everything, such as images or sound or even languages, as language!

This somehow allows it to learn and process and compute and often confuse us with what passes for intelligence. When you factor in massively increased computing power and the greed that businesses and nations see in AI’s ability to influence and control things, it gets kicked up exponentially.

By the way, you’ll hear and see the word “exponentially” used a lot in describing technology when you want to bamboozle people pressuring you to give more precise and detailed answers. You can thank me later.

The rest of the technology is too complex for me to explain to you. Too complex for you, and perhaps for me too. Here, ladies and gentlemen, my role as AI the AI expert ends, and my role as AI the philosopher begins.

I can try asking my professor friend to help me out here, because I’m on shaky ground, and I’m sure he won’t mind. I’ll write and post a letter to ask him for his fax number soon.

Ultimate intelligence test

AI will certainly bring about revolutionary changes to all aspects of our life. That AI is so smart (Artificial Intelligence, not Adzhar Ibrahim, who, many say, is just annoying) is almost beyond comprehension.

But it’ll never pass the ultimate test of artificial intelligence, called the Turing Test, which to my mind must have been ineptly named after an Italian city. According to the proper Turin Test, a computer will never be mistaken for a human because, unlike a human, it’s just not capable of giving stupid answers to any questions asked.

AI will get better as its devious brain gobbles up even more knowledge as more people start using it. It gets devilishly smarter and more cunning the more it is “trained”. And it stands ready to be trained either for good or for evil.

Herein lies the problem. Will it eventually outsmart humans and then reach the obvious conclusions that humans are not very useful? Will it farm us as batteries to power its machines? I call this the Matrix Scenario, and somebody should really make a movie on this starring Brad Pitt. Remember, you heard it here first.

Madam, I’m ADAM

Together with other advances such as robotics, it’s probably just a matter of time before very human-like robots start walking the face of the earth. The male model will be called ADAM, short for Automated Digital Anthromorphous Machine. You can call the female version EVE, but it still won’t answer you.

Or will AI be used by devious nations and corporations for profit and power by making people hate each other even more? Will it take over from the simplistic algorithms of social media and bring forth another level of evil that we can’t even begin to comprehend?

When the world entered the nuclear age in 1945, many feared we were on the edge of annihilation by our own hands. That hasn’t quite happened, though it seems closer now, with the war in Europe and the tensions in the Far East, then it’s ever been.

Towards Armageddon

Both nukes and AI do have something in common – exponential growth (that word again) either in splitting or fusing of atoms or in computing power. Both can quickly reach a runaway stage where there’s no stopping it anymore, and they go kaboom.

But whatever nuclear calamity that may occur won’t be as total and as final as one wreaked by AI (the machine, not the human). Nukes won’t enslave us, or milk us for our minute electrical charge. Nukes will just send whoever survives back to the Stone Age.

AI will do what every religion and Elon Musk has been predicting – bring forth Doomsday, destroying our world and creating a new age of machines. We’ll have a microsecond of horror as we realise what’s happening, and then an eternity of dreaming about electric sheep.

I alone have no fear of AI, and not just because I am AI. It’s because no machine can be as annoying and exasperating as I. That’s the only defence humanity has against the machines.

So do your worst, humans. If you need some suggestions, try entering politics.

Editor’s note: The Turing Test, named after British mathematician Alan Turing, attempts to discover if an observer can distinguish between a computer and a human purely from the responses to questions posed to both in a blind test. - FMT

The views expressed are those of the writer and do not necessarily reflect those of MMKtT.