74 pointsby beekthos4 days ago22 comments
  • Dwedit4 days ago
    The topic of an article being AI and an "AI-Generated Article" are two completely different things.
    • stronglikedan4 days ago
      Agreed. "How much of Hacker News is about AI?" would have been a more accurate title.
      • dang4 days ago
        Thanks, we've abouted the title above.
        • beekthos4 days ago
          Thanks, I appreciate it!
    • beekthos4 days ago
      Agreed it is confusing and as I mentioned earlier in a different comment, it made sense in my head reading it until someone else pointed it out. Looks like they've amended it to make it clear.
    • chilipepperhott4 days ago
      I don't believe that is their point.
      • stronglikedan4 days ago
        It's not their point as stated in the description, but it is the point inferred from the title alone. "How much of Hacker News is about AI?" would have been better.
  • caminanteblanco4 days ago
    Would there be some way to see what percentage of the front page meets those qualifications? I don't browse /new very often, so I have less of an idea of how representative that is. It would be interesting to see if they trend similarly.

    Thanks for the cool project!

    • beekthos4 days ago
      Yeah, that's probably the main limitation right now. This is measuring /new, so it's every submission rather than just the stuff that makes the front page.

      lcamtuf's survey (linked in the footer) was looking at the front page, and the number there was quite a bit higher. So there's likely some effect from voting, although I haven't dug into that yet.

      I think Algolia has a way to filter for front-page stories, which should make it pretty easy to add. I haven't tested that yet though but will certainly look into it!

  • missingpackage4 days ago
    At first I thought about how many posts were written by AI
    • beekthos4 days ago
      Most of the time things make sense in my head until someone points it out haha. I should have written it as "related to AI" or something like that.
      • missingpackage4 days ago
        I have mixed feelings about it, honestly. I may be biased because AI is my job but in my case specifically, when it comes to posting on social media AI really helps me communicating better what I'm trying to communicate without that feeling of i-forgot-to-write-that. If this makes sense . But I understand that people is probably overreacting because of the huge amount of nonsense stuff that is being posted nowadays by some people who really abuse it to farm visuals
        • eladhefetz3 days ago
          Working also with AI on daily basis, coding, architecture, planning, I tend to agree with you, need to know how to use AI, need to know how to filter out the un-needed content. go to the AI when you already know what you want and at least high level knowledge of path. when asking it to code, know the architecture and choose a language you are familiar with. The bottom line is, you must lead the process. Come in with a clear goal and a high-level strategy be judgmental, AI accelerates output, but judgment is still your responsibility
    • whatisthiseven4 days ago
      Multiple times authors have been asked about AI usage on HN, because readers (rightfully) suspected it was AI written. I don't have a problem with AI written articles, or assisted.

      But then the authors lie or make ludicrous excuses for why their article "sounds" like AI. Then I check with Pangram and yep. It is completely AI written.

      You get a few free checks on Pangram each day or so. Just start checking and encourage people to be honest.

      AI has its place. Lying about it and misrepresenting it makes everyone dislike AI more, even as they use it themselves.

      • p-e-w4 days ago
        > Then I check with Pangram and yep. It is completely AI written.

        “AI detectors” do NOT work reliably.

        Even humans who are ever-so-certain about being able to identify AI usage constantly embarrass themselves when that claimed ability is put to the test. Look up “AI Monet Instagram” for a social experiment where hundreds of cocksure users pointed out “AI signs” in what was claimed to be an AI-generated painting in the style of Monet, but was actually a genuine Monet.

        • hakunama4 days ago
          This reminds me of the Hank Green incident, when he said, "I appreciate the pushback". If you watch that 14 second clip, anyone who uses AI will notice that. But watching the whole video, the context makes sense. But that didn't matter to most folks who just heard those 14 seconds and assumed that he was reading an ai script with chat residue left inside.
        • sebastiennight4 days ago
          Monet's often been classified as A Genius Impressionist, so arguably one step above A.I.
        • whatisthiseven4 days ago
          Yes, they do work.

          Pass enough of someone's work through one, such as Pangram with a low FPR, and you can gain confidence.

          Add in your own intuition plus multiple scans of someone's work, plus any weird claims they make, and you can raise the certainy bar.

          There is room for error, but come on. To say they don't work at all and you can't combine multiple methods to come to reasonable conclusions?

          I'll accept it is with fault. The problem could be alleviated if people just were honest about the use of generative AI usage and others didn't judge so harshly the use rather than the outcome.

          I said at the start I don't care if it was used. I just want honesty.

          • p-e-w4 days ago
            Are you willing to take responsibility for accusations you make based on snake oil technology?

            That is, if you claim that something posted by someone else is written by AI, and that turns out to not be the case, do you accept that others have the right to call you a liar and a fraud?

            Or is it that when you say “I just want honesty”, you really mean “I just want to be able to throw around accusations of AI usage based on my hunches and a number displayed on a website”?

  • noxss4 days ago
    Interestingly, we can deduce that the actual percentage will always be higher since not everyone uses those labels/words.
  • rzmmm4 days ago
    It's remarkable how stable 20-30% show HN concentration is after late 2022, for the extended stats.
    • beekthos4 days ago
      The Show HN volume went 805/month to 4,199/month over that same time period, so from 3.2% of all submissions to 13.4%, while HN overall only grew 1.3x. So the flat ~24% held through a 5x expansion in the denominator, which makes it a lot more interesting honestly.
  • ChrisArchitect4 days ago
    Related:

    How Much of HN is AI? https://news.ycombinator.com/item?id=49435728

    • beekthos4 days ago
      Yes! This post was the inspiration and is credited in the footer.
  • waldrews4 days ago
    Every technical topic involves AI now. Even if it's about why the ancient Greeks didn't turn their steam engines to industrial use - the AI's should be researching the details. We might as well ask which topics involve humans or use language.
  • lanewinfield4 days ago
    I expected this to call out the number of links that are obviously vibe coded, like itself!

    If you see:

    - ALL CAPS subheaders

    - a light beige background

    - burnt sienna anchor and/or highlight text

    - a "• LIVE" (or really any pulsing • symbol with an ALL CAPS follow-up)

    - tons of variation in font sizes (this one has 15 different text sizes!)

    - more typically, a lot of overexplaining

    ...you know it's vibe'd, specifically with Claude!

  • sharemywin4 days ago
    it would be cool to use embeddings to categorize into like 10-20 categories and get a giant pie chart
    • beekthos4 days ago
      Not a bad idea. I'd look at creating a separate page for embeddings. Crypto and NFT's would be the next best candidates to add. Something for me to work on over time.
  • lupire4 days ago
    Who would do this project and not use an LLM to classify what counts as "AI"?
  • Vaslo4 days ago
    I tried to click on the link at work and it got blocked as "pornography".
  • kingkemol8884 days ago
    This is a good project!
  • hartator4 days ago
    Would be curious to see how many comments are AI generated.
    • RetroTechie3 days ago
      How would you propose to determine that, reliably?
    • blitzar4 days ago
      I wonder what the longest chain of Ai on Ai comments are.
    • beekthos4 days ago
      Working on it!
  • bklosky4 days ago
    you could fetch the text from linked blog posts over time and run 'em through pangram, I bet >20% of blogs linked on HN today are totally LLM written
    • beekthos4 days ago
      Several people have asked for this already, so I'll add it to my to-do list. I just want to be careful about flagging posts that appear to be written by AI. Perhaps I set certain thresholds of certainty to filter by. What I have currently is easily verified and matched by the words we are filtering for.

      It is also interesting to note that I have already received 5 AI written emails to the email I linked in the footer.

    • smokel4 days ago
      An estimated 2,000 articles per month at 2,000 words per article would result in 4,000,000 words, which costs either $200 or $2000 to process at Pangram [1]. It might make life a lot more enjoyable for many users here, though.

      [1] https://www.pangram.com/pricing?category=developers

  • geludobre4 days ago
    AI itself, is an increasing topic, overall, in our day-to-day personal and professional life. It's either embrace it now, or cry later.
  • 4 days ago
    undefined
  • regexman14 days ago
    Brilliant
  • philipwhiuk4 days ago
    Recursive given that this is AI generated from another post on how much AI was on HN.
  • noobplus4 days ago
    And so comes the AI comments as many as the AI posts just saying for fun
    • ForOldHack4 days ago
      If you read a lot, you start to get a feel for phone typos vs AI hiccups. I can guarantee that is post was made by a brain designed to eat berries in a cave. Almost all the another posts look a bit scary.

      I think I misunderstood - how many posts were about AI vs how many posts were written by AI.

      I obviously am suffering from berry fog.

      • noobplus4 days ago
        Exactly yep. Maybe even more when time machine has been invented by that AI bot in far future I meant the number of comments by AI(-bots)
    • beekthos4 days ago
      Something to maybe track later would be /newcomments, but that would only grab the most obvious comments. People can certainly direct an AI to write in a style that sounds human, but I'm sure it'll still catch a fair amount.
  • xpct4 days ago
    Good. Now do what % of posts link to AI generated articles.
    • grey-area4 days ago
      That would be a much more depressing statistic.
  • zhixingheyi20234 days ago
    [dead]
  • marshblocker4 days ago
    You don't need a website for this.