you are viewing a single comment's thread
view the rest of the comments
[+] -27 points 4 days ago (15 children)

The irony of writing that post with AI... they couldn't write it in their own words but had to use claude for it?

  • source
  • hideshow 15 child comments
  • [–] 20 points 4 days ago* (14 children)

    do you have a base for that accusation?

  • source
  • parent
  • hideshow 14 child comments
  • [+] -6 points 4 days ago* (last edited 4 days ago) (13 children)

    Thats not an accusation, thats the business model

    So we deleted everything. Every product, Every photo. Legally, that may change nothing. Emotionally, it did.

    Both of these are pure LLMisms. If you told claude your story and asked it to make a infographic picture to share on social media you would get really similar writing.

    I use AI a lot at work and have a hypersensitive sense for it. Everyone in this community is LLM-deaf because the majority of members don't use it enough to recognize it's default voice in something.

    The top and bottom paragraph might be their own words its too short for me to tell.

    It leads to hilarious situations like this where users unironically share LLM generated content (in part or in whole) and when I point it out I get downvote bombed. It's doubly ironic when users shame me for calling it out to protect the AI...

    So sometimes I'll argue with users here with claude on default settings, specifically prompting it to be obvious that it's claude but not too obvious. I've never been called on it.

    There's also this:

    People who frequently use ChatGPT for writing tasks are accurate and robust detectors of AI-generated text (Russell et al., ACL 2025)

    In Russel et al's study just 5 "expert" users of AI could predict AI generated text with a majority vote. They had a 299/300 accuracy far exceeding existing AI detectors or the general public.

    There are also studies showing if you don't use AI a lot you don't recognize it. Shocker...

  • source
  • parent
  • hideshow 13 child comments
  • [–] 10 points 4 days ago

    LLMs overweight the style of this particular context of short explanatory posts (not quite long enough for a full blog post but generally longer than a Facebook status or a tweet) for public consumption. LinkedIn is where a lot of this type of post lives, and LLMs tend to borrow from those conventions and extend them outside of that context.

    But that stuff started there for a reason — em dash clauses are an efficient method to give a broad statement and then elaborate/explain in detail after, or by giving an example after.

    • Bullet points are an easy way to visually organize things into lists
    • Bullet points also are easy to textually encode in markdown, which is a friendly format for both training data and LLM output, because of the way LLMs work

    It's not just that the "it's not X, it's Y" sentence structure comes from LLMs; rather, it becomes some kind of logical negative space, that complements the other type of em dash explainer — the general then specific elaboration.

    Many of us learned to write in specific mechanical styles for corporate communications. Those rules aren't that complex, so it stands to reason that those rules end up getting incorporated into how machines generate text, too. But most of us have been doing it for a while, and oscillate between avoiding sounding like LLMs and defiantly using those conventions and saying stuff like why should I change, he's the one who sucks.

    No, I didn't quit using em dashes, bullets, or "it's not X" sentences. But I do use them more judiciously. And because of that, I recognize when they are still absolutely worth the overhead of people accusing the text of being LLM-written.

  • source
  • parent
  • load more comments (12 replies)
  • load more comments (12 replies)
  • load more comments (12 replies)