3 ms·

My ChatGPT meter tells me that it feels like this was written by AI: Notice the excessive use of the em dash throughout the article combined with the sycopha

My ChatGPT meter tells me that it *feels* like this was written by AI: Notice the excessive use of the em dash throughout the article combined with the sycophantic contrasts ChatGPT loves to use so much: "It wasn’t intimacy. It wasn’t mutuality. It was access to simulation — clean, fast and frictionless." A sad state of affairs if the NYT is letting AI-slop through.

voskresenie 1y ago
Let me guess—you watched a tiktok video telling you that em-dashes are a sure sign of AI, and you accepted it without question. Go find a couple NYT articles from 10 years ago and check how many em-dashes they use—the answer might surprise you! (Hint: AI uses a lot of em-dashes because its training data uses a lot of em-dashes. What is its training data?)
audunw 1y ago
Yeah cause it’s completely unimaginable that a professional writer would use em dash /s https://www.nightwater.email/em-dash-ai/ https://www.nightwater.email/em-dash-ai/ I do dread the wave of AI slop coming our way. But honestly what I dread even more is the comments that’s going to second guess every other article and comment, just because they use some character or style that people have decided is a sign of being AI generated. That’s gonna get tedious real fast.
- voskresenie 1y ago
  I'm an amateur composer and it pains me to know that, given my lack of credentials, my work will always be viewed with the suspicion of AI involvement. I version control my scores for this reason, at the very least I can prove that it took me weeks, months, or sometimes years of effort and revisions to compose something, even if I can't prove that each revision wasn't just the output of an AI. It was already bad enough worrying that a melody or sequence I wrote is too similar to some other existing work and judged to be plagiarism, but that pales in comparison to AI suspicions.
inkywatcher 1y ago
The dashes thing is really only a tell for high schoolers and obvious illiterates –writers can and do use them. Next we need to teach GPT about interrobangs.
- voskresenie 1y ago
  thank you for saying this. I'm not a writer, just vaguely literate, and I find em-dashes to be extremely useful as a way of approximating conversational speech without a string of comma splices. Additionally, the observation about AI using dashes really only applies (if it applies at all) to informal text conversation, not published articles. Here's a random NYT article from 2014 that uses 4 em-dashes within the first 3 paragraphs. https://www.nytimes.com/2014/12/21/fashion/the-year-of-taylor-swift.html https://www.nytimes.com/2014/12/21/fashion/the-year-of-taylo... It seems like everyone watched some video essay about how em-dashes are a sure sign of AI, and they're just parroting it without question. The thing that apparently goes over people's heads is that the reason AI tends to use a lot of em-dashes is because the text in their training data uses a lot of em-dashes, and the training data is largely published articles and books, so it's a terrible heuristic for whether a published article or book was written using AI. The thing that actually is somewhat more telling is that many people will use hyphens instead of dashes, eg on a computer, I typically type '--' instead of an em-dash, partly because my xcompose setup is inconsistent, partly because I write most of my text in editors that use monospaced fonts and the distinction between -, – , and — is extremely subtle in most fixed-width fonts, for obvious reasons. But on macs and many phones, as well as in google docs and similar, by default a hyphen will be autocorrected to an en- or em-dash depending on context, so it's not really a tell that the entire thing was written with AI, just that there was possibly non-human involvement. But also, a lot of people actually just know how to type and use dashes. It's not really that hard.
ashwinsundar 1y ago
Your ChatGPT meter is probably broken
carabiner 1y ago
I actually totally agree and was going to post this until I saw your comment. The writing is strikingly ChatGPT style. So many commas, weird phrase lists (especially with 3 items): > No shrinking. No waiting. No apologizing. > slowness, curiosity, accountability > Avoidance. Exhaustion. Disrepair. > It wasn’t intimacy. It wasn’t mutuality. It was access to simulation.
- liquidki 1y ago
  To be sure, this is also taught in writing workshops, speaking workshops, comedy and improv as well, the rule of three: https://en.wikipedia.org/wiki/Rule_of_three_(writing) https://en.wikipedia.org/wiki/Rule_of_three_(writing)
  - carabiner 1y ago
    Yes and chatgpt is incredibly heavy handed with it. This is not how most NYT articles are written.

My ChatGPT meter tells me that it *feels* like this was written by AI: Notice the excessive use of the em dash throughout the article combined with the sycopha

voskresenie 1y ago

Let me guess—you watched a tiktok video telling you that em-dashes are a sure sign of AI, and you accepted it without question. Go find a couple NYT articles from 10 years ago and check how many em-dashes they use—the answer might surprise you! (Hint: AI uses a lot of em-dashes because its training data uses a lot of em-dashes. What is its training data?)

audunw 1y ago

Yeah cause it’s completely unimaginable that a professional writer would use em dash /s https://www.nightwater.email/em-dash-ai/ https://www.nightwater.email/em-dash-ai/ I do dread the wave of AI slop coming our way. But honestly what I dread even more is the comments that’s going to second guess every other article and comment, just because they use some character or style that people have decided is a sign of being AI generated. That’s gonna get tedious real fast.

voskresenie 1y ago
I'm an amateur composer and it pains me to know that, given my lack of credentials, my work will always be viewed with the suspicion of AI involvement. I version control my scores for this reason, at the very least I can prove that it took me weeks, months, or sometimes years of effort and revisions to compose something, even if I can't prove that each revision wasn't just the output of an AI. It was already bad enough worrying that a melody or sequence I wrote is too similar to some other existing work and judged to be plagiarism, but that pales in comparison to AI suspicions.

inkywatcher 1y ago

The dashes thing is really only a tell for high schoolers and obvious illiterates –writers can and do use them. Next we need to teach GPT about interrobangs.

voskresenie 1y ago
thank you for saying this. I'm not a writer, just vaguely literate, and I find em-dashes to be extremely useful as a way of approximating conversational speech without a string of comma splices. Additionally, the observation about AI using dashes really only applies (if it applies at all) to informal text conversation, not published articles. Here's a random NYT article from 2014 that uses 4 em-dashes within the first 3 paragraphs. https://www.nytimes.com/2014/12/21/fashion/the-year-of-taylor-swift.html https://www.nytimes.com/2014/12/21/fashion/the-year-of-taylo... It seems like everyone watched some video essay about how em-dashes are a sure sign of AI, and they're just parroting it without question. The thing that apparently goes over people's heads is that the reason AI tends to use a lot of em-dashes is because the text in their training data uses a lot of em-dashes, and the training data is largely published articles and books, so it's a terrible heuristic for whether a published article or book was written using AI. The thing that actually is somewhat more telling is that many people will use hyphens instead of dashes, eg on a computer, I typically type '--' instead of an em-dash, partly because my xcompose setup is inconsistent, partly because I write most of my text in editors that use monospaced fonts and the distinction between -, – , and — is extremely subtle in most fixed-width fonts, for obvious reasons. But on macs and many phones, as well as in google docs and similar, by default a hyphen will be autocorrected to an en- or em-dash depending on context, so it's not really a tell that the entire thing was written with AI, just that there was possibly non-human involvement. But also, a lot of people actually just know how to type and use dashes. It's not really that hard.

ashwinsundar 1y ago

Your ChatGPT meter is probably broken

carabiner 1y ago

I actually totally agree and was going to post this until I saw your comment. The writing is strikingly ChatGPT style. So many commas, weird phrase lists (especially with 3 items): > No shrinking. No waiting. No apologizing. > slowness, curiosity, accountability > Avoidance. Exhaustion. Disrepair. > It wasn’t intimacy. It wasn’t mutuality. It was access to simulation.

liquidki 1y ago
To be sure, this is also taught in writing workshops, speaking workshops, comedy and improv as well, the rule of three: https://en.wikipedia.org/wiki/Rule_of_three_(writing) https://en.wikipedia.org/wiki/Rule_of_three_(writing)
- carabiner 1y ago
  Yes and chatgpt is incredibly heavy handed with it. This is not how most NYT articles are written.