Having a Few Laughs with ChatGPT

See the latest issue of Briefings at newsstands or read in our new format here.

By: Jonathan Dahl, Chief Content Officer

It’s been half a year now, and the world is still going gaga over ChatGPT and its various competitors. And why not, given its ability to do everything from instant research to designing new corporate operations? But for my part, I had a more humble goal for the tool: Could my new AI friend write my Briefings column for me? Five hundred words, please.

“Of course!” it eagerly responded, without a single complaint about how I was making it do my job. I figured a good topic would be the future of artificial intelligence. In 20 seconds flat, ChatGPT rattled off some “very exciting” opportunities and advancements it could offer in the fields of medicine, finance, and transportation. Almost breathlessly, it informed me that “the most significant potential for AI lies in its ability to transform our daily lives.”

But I wanted more than information. I wanted to see some writing chops. “Hey ChatGPT, can you write my column with some humor?”

“Of course!”

And here, I think, is where things get interesting. I’ll let the AI speak for itself.

“The future of artificial intelligence (AI) is so bright, you might want to wear some shades. It’s amazing how much progress we’ve made in just a few short years. Pretty soon, we’ll have AI robots walking around doing our laundry and cooking us breakfast. Okay, maybe that’s a bit of an exaggeration, but you get the point.” Actually, I didn’t. It went on:

“The finance industry is also getting a taste of the AI revolution. With AI algorithms analyzing data faster than a cheetah chasing its prey, investment decisions and risk management have never been easier. And let’s not forget about virtual assistants and chatbots—say goodbye to those cringeworthy small-talk conversations with human bankers.”

Clearly, as a writer, ChatGPT could probably work on its similes (at one point, it used one of the tiredest—“as punctual as a Swiss watch”). Sensitivities were an issue too; I’m not sure everyone finds small talk with bankers “cringeworthy” and/or would want to offend them. Interestingly, I think my command to add humor actually provoked the AI to hype itself. It informed me, for example, that self-driving automobiles are not only a reality right now—but that also “soon we’ll be napping in the backseat while our car takes us to work.”

In conclusion (which, incidentally, was how ChatGPT ended each column it wrote for me), it’s hard to not be impressed with how quickly this tool can provide some nifty research. As for the writing, a little, very human research of my own discovered that a hot new job these days are for “prompt engineers”—the people who know how to ask AI the right questions. I suspects the prompts for my columns were all wrong.

In other words, human error—once again.