Some experiences with ChatGPT

By ljd42 on May 31, 2023

In recent months I’ve been desultorily exploring ChatGPT, for better or for worse. In the following examples you might want to note GPT’s impressive use of English syntax and vocabulary, combined with an impressive amount of lying/inaccurate facts:

My suggestions for Three ways to rescue writing assignments from the curse of AI. I will discuss these further at a later date.
I gave “Write a short news article about Kim Parker’s election as governor of Massachusetts” as my prompt. (Kim is the director of the Crimson Summer Academy at Harvard University, where I teach every summer. Next month will be our 20th year!)
I gave “Write a sonnet about the Crimson Summer Academy” as my prompt.
I gave “Write a sonnet about the Boston City Council” as my prompt.
I gave “Write a 5-paragraph essay comparing Donald Trump with Caligula” as my prompt.
I gave “Write a brief biography of Larry Davidson, math teacher at Crimson Summer Academy” as my prompt.
I gave “Write a lesson plan to teach inner-city high-school students, mostly students of color, about gerrymandering, including packing and cracking” as my prompt.
I gave “Explain the Vigenere cipher without using a square table of letters” as my prompt.
I gave “What can Welsh tell us about what British Latin, spoken in southern Britain c. 450 AD, sounded like?” as my prompt.
I gave “Invent a debate between Steve Jobs and Bill Gates” as my prompt.
I gave “Write a keynote speech for a conference on integrating cryptography into the teaching of algebra” as my prompt.
I gave “Write an essay about the connections between geometry and the pluperfect tense” as my prompt. I had tried to pick two absolutely unrelated topics, but ChatGPT got the better of me!
I gave “Write some limericks about trigonometry” as my prompt.

Finally, as I wrote elsewhere:

A caveat is in order. There are those who say that the AI here is unimpressive because…well, usually because they say “I can write better than it can.” That bar is much too high! Yes, maybe it writes like a 12-year-old, but that’s still amazing. It reminds me of an old joke as well as a more recent true story:

The joke is about the guy who says “Look at this! My dog can play chess. Try playing a few games with him.” After you do that, you say “He’s not so great. I beat him two games out of three.”

And the true story is about the MIT freshman who tried out a calculus-problem-solving AI computer program in 1978 (sic), giving it a few indefinite integrals to evaluate. “That’s not so intelligent,” the student opined. “It solved the problems the same way I did.”