In a way, this comment precisely demonstrates the kind of vitriol responses that the article talks about -- witch hunt without evidence, going with the trend, herd mentality.
Really? A comment in another thread says Pangram says it's 100% AI. I just checked with the free plan, and it told me 29% AI. What I'm learning here is that Pangram may not be a reliable checker.
Thing is, I also twigged it as heavily LLM written when I read it, so. ¯\_(ツ)_/¯
I checked again with a longer sample—the first eight paragraphs—and, this time, Pangram said 61% human. Hmm. I think your twigging may be more accurate.
Even if so, this would seem to be a positive use of LLMs. Something traumatic happened, and she wants to put her perspective into the public dialog rather than continuing to be a passive subject of the social media industry's two minute hate. I know in such a position I myself would be at a loss for words, with the perfect being the enemy of the good enough.
In a way, this comment precisely demonstrates the kind of vitriol responses that the article talks about -- witch hunt without evidence, going with the trend, herd mentality.
https://fuck-off.ai/panned
Isn't this story pretty much all about jumping to conclusions and pissing on people who you don't know?
I read it and didn’t get that impression.
Perhaps you could endow your claim with some evidence.
Amused by a claim about writing composition that uses “clearly mostly” back to back.
That's a legitimate usage of both words.
there are higher levels of writing than "grammatically correct"...
we don't dismiss claims of poorly written code with "but it compiles!"
Hmm. Pangram AI detection says the piece is “100% human written.”
Really? A comment in another thread says Pangram says it's 100% AI. I just checked with the free plan, and it told me 29% AI. What I'm learning here is that Pangram may not be a reliable checker.
Thing is, I also twigged it as heavily LLM written when I read it, so. ¯\_(ツ)_/¯
I got the vibe that maybe an LLM was used for editing but the stench wasn't overwhelming.
I checked again with a longer sample—the first eight paragraphs—and, this time, Pangram said 61% human. Hmm. I think your twigging may be more accurate.
Based on what?
Even if so, this would seem to be a positive use of LLMs. Something traumatic happened, and she wants to put her perspective into the public dialog rather than continuing to be a passive subject of the social media industry's two minute hate. I know in such a position I myself would be at a loss for words, with the perfect being the enemy of the good enough.
[dead]