This essay is--oddly--both threadbare and meandering, and has turns of phrase that look like they were kept from an earlier, tighter version, such as the second sentence's "may have [...] but by".
Now that people know what ai slop is, they start having higher expectations from prose because vacuous articles like these might be misconstrued as slop
What a fun little essay. It takes a lot of skill to write such a nonlinear story of ideas and history, and to have the structure and the actual point emerge gradually like this. It also takes some attention and some faith on the part of the reader, and it sounds like a few people here got frustrated that it doesn't feel like a normal essay that tells you upfront where it's going and why.
One commenter saw the long second sentence and assumed that the author had messed up somehow, instead of clocking the heads-up that this is piece is going to be a bit unusual and demanding. We're slowly getting used to how LLMs keep every sentence short enough to digest without much effort.
An LLM could never write this, as others here have pointed out, at least not without some very involved, nonstandard prompting. I think it's going to be important for our brains to keep reading stuff like this that is structurally unique. It almost reads like a miniature Benjamin Labutut story.
(Ok ok ... the real quote is actually: "The Best Way to Predict the Future is to Invent It", and I actually think he also had that quote from someone else, earlier, but I forgot the name.)
The article concludes, stating that LLMs are the “apotheosis of knowledge stripped of animating spirit and reduced to mere utility.” Maybe, but LLMs have far more utility than this directionless essay.
This essay is--oddly--both threadbare and meandering, and has turns of phrase that look like they were kept from an earlier, tighter version, such as the second sentence's "may have [...] but by".
I read some AI slop earlier today and then read this. The AI slop was better than whatever this is.
You might not be the target audience for this, and that's ok.
Now that people know what ai slop is, they start having higher expectations from prose because vacuous articles like these might be misconstrued as slop
What a fun little essay. It takes a lot of skill to write such a nonlinear story of ideas and history, and to have the structure and the actual point emerge gradually like this. It also takes some attention and some faith on the part of the reader, and it sounds like a few people here got frustrated that it doesn't feel like a normal essay that tells you upfront where it's going and why.
One commenter saw the long second sentence and assumed that the author had messed up somehow, instead of clocking the heads-up that this is piece is going to be a bit unusual and demanding. We're slowly getting used to how LLMs keep every sentence short enough to digest without much effort.
An LLM could never write this, as others here have pointed out, at least not without some very involved, nonstandard prompting. I think it's going to be important for our brains to keep reading stuff like this that is structurally unique. It almost reads like a miniature Benjamin Labutut story.
Alan Kay.
(Ok ok ... the real quote is actually: "The Best Way to Predict the Future is to Invent It", and I actually think he also had that quote from someone else, earlier, but I forgot the name.)
https://quoteinvestigator.com/2012/09/27/invent-the-future/
Thanks. I just went down a little rabbit hole into Dennis Gabor and then the book Cybernetics. Fun. :-)
The article concludes, stating that LLMs are the “apotheosis of knowledge stripped of animating spirit and reduced to mere utility.” Maybe, but LLMs have far more utility than this directionless essay.