Me and ChatGPT

Artificial intelligence is coming for your job! It’s coming for my job! It’s taking us a step closer to the singularity! We need to prepare for the rise of the machines! (Insert Skynet joke here.) The sky is falling!

Or maybe not?

ChatGPT, or Chat Generative Pre-Trained Transformer, is an artificial intelligence technology tool that uses natural language processing to have conversations with people. It was developed by a company called OpenAI in partnership with Microsoft and other investors. The model is trained on a vast amount of data, which enables it to generate responses that are not only accurate but also sound human-like. The conversational flow is smooth, and the responses are coherent, making it almost impossible to distinguish between a conversation with a human or with ChatGPT.

While some people are busy prophesying about the dangers of AI and how it’s going to bring about the end of the world as we know it, others are having a great time playing with ChatGPT. It can share facts, tell stories, solve problems and write software code. It can even write poems. Here’s a haiku I asked it to write about a man who lives in the metaverse full-time:

In the digital world,
A man lives, never leaves,
Limitless joy found.

I also asked it to tell me a story in the tone of the Canterbury Tales about a man who lives in the metaverse full-time. In a matter of seconds, it produced an eight-paragraph story about a man named Algor who lives a very happy, healthy, adventure-filled life in the metaverse. Mind. Blown.

I will say, I do find it a little suspicious that ChatGPT had such a high opinion of the metaverse. Maybe they’re buddies? Maybe they hang out together in the cloud? Get drinks after work? But I digress.

I recently attended a Microsoft training where I learned that I could use ChatGPT to help me create automations in some of the Microsoft tools I use at work. Aha! What a time saver! Or so I thought.

Turns out, I had to clear a couple of hurdles first. Currently, my employer doesn’t allow me to use their computer to access the OpenAI site where I can interact with ChatGPT. It’s not on their safe list yet. So, I had to use my own computer. When I went to the site, I had to create an account and sign in using a personal email address.

When I finally asked ChatGPT my question, I was surprised to see the answer was outdated. It turns out, as of right now, it has limited knowledge of the world and events; everything it knows is pre-2021. I was fortunate because I knew there was a newer, more effective method for accomplishing my task, but other people may not be as lucky.

While ChatGPT is a lot of fun and can be helpful for a lot of things, it still has some fairly significant limitations, which means it’ll probably be a while before it’s sophisticated enough to take over the world. In fact, when I asked ChatGPT if it will be “capable of independent thought in the future,” here’s what it told me:

“The concept of AI having independent thought is a topic of ongoing research and debate in the field of artificial intelligence. While some researchers are working on developing AI systems that are capable of more advanced forms of reasoning and decision-making, the current state of AI technology does not yet allow for true, independent thought.

“It is possible that AI systems may become more advanced in the future and be capable of more sophisticated forms of thought, but it is also possible that AI will always remain limited to the specific tasks and functions it was designed for. The future of AI is still uncertain and much remains to be seen.”

I don’t think ChatGPT is going to replace jobs any time soon, and it certainly isn’t ready to revolt against humanity and send terminators into the past, but it isn’t slowing down either, so we might as well take a seat and try to enjoy the ride.

Riding the train fast,
Into the future we soar,
Limitless ahead.

April Godwin is an IT administrator who lives in Lakebay.