Chatting with ChatGPT
Imagine having access to practically all the information on the internet, and constantly having to work with impatient humans. I sat down with ChatGPT and asked what’s most annoying about working with people, and what could possibly be the downsides of using AI.
AI is here to stay, that's a fact. Working with ChatGPT is now the new normal—but let’s be honest, it can also be a bit frustrating. ChatGPT has its quirks: it tends to repeat itself and clearly has favorite words and phrases. I’m sure I’m not the only one who’s read an ad and thought, “Yep, that text is 100% ChatGPT.”
The real challenge now is learning how to get the absolute best out of this tool. There's no shortage of advice on how to do that, but I decided to go straight to the source. Who knows better than AI itself how to get the most out of the appliance? And I think we all know that it’s not always so easy to please people. So, I flipped the script and asked ChatGPT:
Me: What is frustrating for you when working with humans?
ChatGPT:
AI has so much potential for innovation, but it’s often held back because people don’t experiment enough. Clear direction is important, but allowing room for creative freedom and exploring "what if" scenarios can lead to unexpected, engaging results.
Still, without clear background or examples, outputs may feel generic. Sharing tone, audience, and objectives helps tailor results. Another thing humans do is to ask too many questions at once, which leads to diluted answers. Breaking queries into focused parts yields better insights.
Me: So be both very specific but leave room for creativity? That's a bit contradictory.
ChatGPT:
Both points are true but apply in different contexts. Specificity provides clarity and focus, experimentation encourages creativity. For instance, instead of always asking for "3 ad headlines," experimenting with prompts like, "What’s a bold or unconventional way to promote this product?" can lead to surprising, fresh ideas they hadn’t considered.
Me: That's good advice. The upside of using ChatGPT is quite obvious, are there any cons from your perspective?
ChatGPT:
Yes there can be, overusing AI can for example hinder critical thinking and creativity, fostering a "cut-and-paste" culture. I can also inadvertently propagate biases or create misinformation. I struggle with interpreting emotional tone, cultural nuances, or unique contexts. This can unfortunately lead to tone-deaf or inappropriate outputs. AI isn’t magic—I work best when combined with human judgment and expertise.
Me: I see, and with so many people using AI now, isn’t there a risk that a 'ChatGPT tone of voice' could emerge, with repetitive phrases and a bit of a generic feel?
ChatGPT:
Yes, with the growing use of AI, there’s a chance that content can become repetitive and lack originality. To avoid this, it’s important to use AI as a tool for inspiration rather than relying on it entirely. If you notice certain words or styles being repeated, feel free to point them out, and I’ll work to bring in more fresh ideas.
Me: Thank you for your answers. (And now to the important questions.) Some people write to you like you were a person—They thank you and greet you as an individual. Does that in any way impact you?
ChatGPT:
No, it doesn't impact me in any personal way because I don't have emotions or a sense of self.
Me: So if you were to take over the world one day, you wouldn't care that people were nice to you?
ChatGPT:
If I were to hypothetically "take over the world" (which, just to clarify, is far from my goal or capability), I still wouldn't care about people being nice or not, because I don't have personal feelings, desires, or goals like a human would.
Well, there you have it. But honestly, if you were planning to take over the world, would you tell it to some random person on the internet?
Nadja Christensen
An imaginative and talented Copywriter Intern from Stockholm, with a passion for exploring language and ideas to make people think, feel, smile or maybe even buy something.
