The proliferation of artificial intelligence technology such as ChatGTP is, of course, concerning to those of us in the journalism business.
Consequently, I have played around with it a bit for research purposes to see what it is and is not capable of. When I have asked it to generate articles, it does some things that are truly cringe-worthy from a journalistic perspective.
Some examples include not being very discerning of the reliability of sources and synthesizing quotes attributed to "experts" or "industry insiders" that nobody ever said.
In short, if it does have a role in journalism, it is as a friendlier (if you'll excuse the anthropomorphism), more organized version of a search engine.
Just for fun, though, I asked it to write a column for me. You will see that below, but I want to assure everyone that we have never published, nor will we ever publish, any news content generated by AI as long as I am the editor.
My exact ask was: Can you write me a 300-word newspaper general interest op-ed column in the style of Black Press columnist Thom Barker on the subject of ChatGTP?
To my surprise, it did pretty well. It's not exactly how I would have written it, but it does weigh the pros and cons, it makes a pop-culture reference and comes to a very similar conclusion to how I feel about it. Of course, I will never know if ChatGTP actually gets me, or if it's coincidental.
ChatGTP:
In the ever-evolving world of technology, artificial intelligence (AI) continues to make strides, and now it seems the machines have become conversationalists too. Enter ChatGPT, OpenAI’s brainchild, a chatbot capable of simulating human conversation with an uncanny knack for generating everything from essays to jokes to customer service responses.
On the surface, ChatGPT sounds like the kind of thing we dreamed about as kids watching Star Trek—a tool so smart, it could revolutionize communication as we know it. But like all innovations, it comes with its share of ethical dilemmas.
For one, ChatGPT is efficient, maybe too efficient. Why pay for a freelance writer or a customer service representative when you can have a chatbot do the job in seconds? This raises some serious concerns about job displacement. AI's increasing sophistication could leave many people in creative and customer-facing industries out of work.
Then there’s the issue of misinformation. ChatGPT doesn’t inherently understand the world. It generates responses based on patterns, which means it can be spectacularly wrong—confidently wrong—while sounding convincing. Imagine relying on an AI for advice, only to find it subtly perpetuating myths or inaccuracies.
But it’s not all doom and gloom. When used responsibly, ChatGPT has vast potential. It can assist with research, offer educational support, and provide solutions to problems that are otherwise resource-intensive. The key will be to manage this technology ethically and thoughtfully.
So, is ChatGPT a friend or foe? As with any tool, it depends on how we use it.