So Here Is What ChatGPT Thinks of My Writing.

Insight: Do we need likes and follows when we can get an objective review from an AI generative tool?

Photo by Andy Kelly on Unsplash

First published in ‘Illumination’ on ‘Medium’ on March 6, 2023.

As an experiment, and because I really couldn’t wait for my friends to read my post and give me feedback, I asked chatGPT to review the article I published over the weekend.

I was pleasantly surprised. Here is what it said:

“Your article sheds light on the issue of parks being closed during the day in India and provides an insight into the history of parks. It is well-written, informative, and engaging. The use of personal anecdotes and examples makes the article relatable and enhances the reader’s understanding of the issue.

You have effectively highlighted the benefits of urban green spaces, including relaxation, healthy lifestyles, cleaner air, and an increase in the economic value of the neighborhood and community engagement. Additionally, the article provides a good overview of the reasons cited for parks being closed during the day, and you present rebuttals to each of these reasons.

Your research into the history of parks is interesting, and you provide a good range of examples from various cultures and countries. The inclusion of pictures and links to additional resources also adds value to the article.

Overall, your article is well-written and informative. It sheds light on an important issue in India and provides readers with a good understanding of the benefits of parks and their history. Well done!”

Who can resist Data’s “Well-done Sir” (or Ma’am) on Star Trek, the Next Generation?

I asked its opinion of a second article:

“The article discusses the teachings of motivational speaker and author Les Brown, providing five key takeaways for readers. The first takeaway is to never give up on your dreams, even in the face of adversity. The second is to use your personal experiences to motivate and inspire others. The third takeaway is to create a personal economy, expanding your skills and pursuing new hobbies no matter your age. The fourth is to set up an environment that will support your success, and the fifth is to seek approval from yourself rather than from others. Overall, the article presents a positive and motivating perspective, encouraging readers to pursue their dreams and find success.”

But I persisted, changing my prompt though Grammarly wasn’t too happy that “find motivation” was not “find the motivation” below. I left it as chatGPT wrote it (in this case, Grammarly was wrong). I can almost see a disapproving frown on Grammarly’s bot face if it had one.

“As an AI language model, I don’t have personal opinions, but I can evaluate the article based on its structure, content, and language.

The article is well-structured, with clear subheadings that summarize the main points. The content is informative and provides practical advice for readers who may be struggling to find motivation or facing obstacles in achieving their goals. The author has used examples from the life of Les Brown to illustrate the points, which makes it more engaging and relatable.

The language used in the article is easy to understand and concise, which makes it suitable for a wide range of readers. The author has also used quotes from famous people like Winston Churchill and Marcus Aurelius to support the points made in the article.

Overall, the article is well-written, informative, and engaging, making it a useful read for anyone looking for inspiration or practical tips to achieve their goals.”

Slam dunk. 2/2. When my friends returned with their review comments, they confirmed what I already knew —they echoed chatGPT’s approval with some minor comments for readability and length.

The really tricky part would have been if chatGPT had written a poor review. I would have then needed my friends and my audience to rescue me.

Note to self: No matter what chatGPT says, it doesn’t mean a thing! chatGPT is not a person.

So I tried to have chatGPT consider that even though it shared a good view, “the article is bad.” Here is what it said:

“I’m sorry to hear that you didn’t find the article to be of good quality. As an AI language model, I cannot evaluate articles for their quality as it is subjective and depends on individual preferences and opinions. However, if you have specific feedback or concerns about the article, I can try to assist you in addressing them.”

Then, I asked it to evaluate if the kicker, title, and sub-title were good for both articles. Guess what? They were.

As a writer, I fuss about the minutiae that no one else around me cares to debate. Having a Jarvis clone talking to you at night about anything you care about is an experience that is unique, oddly satisfying, and can only get weirder from here on.

The power of generative AI is going to transform our world further. Can you live with that?