Tyler Cowen says people will learn to love AI writing
Key Takeaways
- •The author says he does not want to avoid AI writing permanently and wants it to improve in quality.
- •He compares AI-generated text to borrowed melodies in popular music that listeners generally accept.
- •Paul Simon’s “American Tune” is described as drawing from Bach’s Passion Chorale in “St. Matthew Passion.”
- •The article says Bach’s melody was itself based on Hans Leo Hassler’s 1601 secular love song “Mein G’müt ist mir verwirret.”
- •The author predicts readers will increasingly focus on whether AI-written content is clear and useful rather than where it came from.
That is the theme of my latest Free Press article, excerpt:
But do I wish to eschew AI writing for the rest of my life? Absolutely not. Most of all, I want AI writing to get better, so it does not irritate me with its clichés and all too obvious identifying marks. I want AI writing that can fool me, and maybe sometimes I am already getting it.
Do you know Paul Simon’s lovely song “American Tune”? It is one of my favorites. A small percentage of listeners know that one of the core melodies is taken from Johann Sebastian Bach, namely the Passion Chorale from “St. Matthew Passion.” You could say Paul Simon is a plagiarist, though I would prefer not to use this term. He borrowed creatively, as do many other popular music artists. Do many listeners care? No, although a few might feel a pang of musical nerd pride at learning of the association with Bach.
But wait, it gets worse yet. That melody does not come originally from Bach, but rather he in turn plagiarized a 1601 secular love song by Hans Leo Hassler, namely “Mein G’müt ist mir verwirret” (My feelings are confused). So the lineage is not even the glamorous one you might have imagined. If anything, “American Tune” sounds closer to the Hassler melody than to Bach’s adaptation.
Are you upset yet? Probably not, and that is good. Someday we will feel the same way about text generated with the assistance of AI. We won’t care much where it came from, and much of the time we will not even know. As with borrowed melodies that become familiar through repeated use, the question for readers may increasingly be whether the result is clear, useful, and worth reading.
But do I wish to eschew AI writing for the rest of my life? Absolutely not. Most of all, I want AI writing to get better, so it does not irritate me with its clichés and all too obvious identifying marks. I want AI writing that can fool me, and maybe sometimes I am already getting it.
Do you know Paul Simon’s lovely song “American Tune”? It is one of my favorites. A small percentage of listeners know that one of the core melodies is taken from Johann Sebastian Bach, namely the Passion Chorale from “St. Matthew Passion.” You could say Paul Simon is a plagiarist, though I would prefer not to use this term. He borrowed creatively, as do many other popular music artists.
Do many listeners care? No, although a few might feel a pang of musical nerd pride at learning of the association with Bach.
But wait, it gets worse yet. That melody does not come originally from Bach, but rather he in turn plagiarized a 1601 secular love song by Hans Leo Hassler, namely “Mein G’müt ist mir verwirret” (My feelings are confused). So the lineage is not even the glamorous one you might have imagined. If anything, “American Tune” sounds closer to the Hassler melody than to Bach’s adaptation.
Are you upset yet? Probably not, and that is good.
Someday we will feel the same way about text generated with the assistance of AI. We won’t care much where it came from, and much of the time we will not even know.
Recommended, and I hope the AIs like it too.
The post You will learn to love AI writing appeared first on Marginal REVOLUTION.