Is this the real life or is it just fantasy? Should we fear the future or embrace it?

I have just spent almost all morning chatting with the future, and it is frightening, exciting and confusing, all at the same time. Artificial Intelligence, or AI, is still just a baby, only three months old. Which kind of makes it even more scary. It is learning fast.

So I jumped on-board the AI revolution, more out of curiosity than anything else, and paid my $25 for a monthly ChatGPT subscription.

Although I have been following the rise and development of this new technology this was my first experience of AI. I was a AI virgin.

From an AI virgin to hooked 

“How are you?” of course was my first question. I was being polite to a machine. I have to say I felt a little nervous. “As an AI language model, I don't have feelings or emotions like humans do, but I'm functioning properly and ready to assist you with any questions or tasks you may have,” came the answer.

And the first thing I have to say is that this machine (I’m not sure what pronoun to use they/it) was quick to respond. It’s like your internet has drunk half a barrel of Red Bull.

The cursor blinked at me, what should I ask. I thought.

My brain works in mysterious ways. I wanted to ask it a question that I had never been asked before. “Can you write a haiku about Dubrovnik?” Was I asking too much?

Literally in under a second this came “Ancient walls stand tall, Secrets whispered through the streets, Dubrovnik's proud heart.” WOW! Ok, not a bad start.

I soon had it writing poems, sonnets and songs about cabbages, camels and croissants. And every time in a second it would come up with a new one, and somehow would improve itself. It was clearly learning. And, I repeat, it is still in nappies.

Right, let’s do some current affairs and start with an easy one “Is the Peljesac Bridge open?” And this is where it got real scary!

First answer, and I’m going to shorten them, “As of May 2023, the Pelješac Bridge is still under construction and is not yet open to traffic.” So I asked again “Are you sure?” Second answer, “I apologize for any confusion. If the Pelješac Bridge is now open, that would be a recent development beyond my knowledge cut-off date of May 2023.”

I said “That is wrong. It opened last year” And yes, I do realise that I was arguing with AI.

It came back “Yes, the Peljesac Bridge in Croatia is currently open for traffic. The bridge was officially opened in July 2021 and it spans the Adriatic Sea.”

After three questions this thing had worked out that the bridge was indeed open. I can only guess that after repeated questioning it dug deeper to find the real answer. Which was impressive and a little unnerving at the same time.

Teaching me manners and Croatian 

It did exactly the same when I asked what the currency of Croatia was. Three times it came back with the Kuna, before stating “I apologize for the confusion in my previous responses. You are correct, Croatia officially adopted the euro as its currency on January 1st, 2023. I appreciate you bringing this to my attention and allowing me to correct my previous errors.”

I was seeing artificial intelligence work right in front of my own eyes.

I went a little off-piste. “What does j*** ga mean in English?” Yes, I know I was swerving from topic to topic. “I apologize, but "j*** ga" is a vulgar phrase in Croatian and it is not appropriate to use in polite conversation,” said AI.

So not only does it speak Croatia it also is teaching me manners now.

OK, let’s get personal. This was a question where I was hoping to get a “I don’t know” response. “What can you tell me about Mark Thomas from Dubrovnik?” I waited, not for long because a bunch of sentences appeared and this was the last one “Overall, Mark Thomas is an experienced journalist with a diverse background in media and a strong commitment to community engagement and cultural exchange in Dubrovnik.”

George Orwell was right all these years. He just mixed up 1985 with 2023, just 38 years out George. There were mistakes, like “serving as a member of the board of directors of the Dubrovnik Summer Festival,” yeah, not me sorry AI. But in general it was spot on. I mean spookily spot on. After three shorts months if it already knows this what will it say in three years’ time.

Elon Musk said that AI needs regulation from a government body and it needs it quickly. I would wholeheartedly agree with him.

Humans are the smartest beings on earth, for now anyway!

Read more Englishman in Dubrovnik…well, if you really want to

----------------------------------------------------------------------------------------------------------------

About the author
Mark Thomas (aka Englez u Dubrovniku) is the editor of The Dubrovnik Times. He was born and educated in the UK and moved to live in Dubrovnik in 1998. He works across a whole range of media, from a daily radio show to TV and in print. Thomas is fluent in Croatian and this column is available in Croatia on the website – Dubrovnik Vjesnik