• Zagorath@aussie.zone
    link
    fedilink
    English
    arrow-up
    19
    ·
    edit-2
    1 month ago
    Transcription

    A post on Threads, consisting first of text, and then of a screenshot of two Tweets, each a reply to the one after it.

    By “georgehtakei”:

    This was an IQ test.

    By “Catturd ™” @catturd2:

    😂😂😂 we warned you.

    By “Polymarket” @Polymarket:

    BREAKING: Zohran Mamdani to require all New York elementary school students to learn Arabic numerals.

    • egrets@lemmy.world
      link
      fedilink
      arrow-up
      7
      ·
      1 month ago
      Slight correction (and perhaps unimportant)

      Only George Takei’s post was on Threads. He’s sharing a screenshot of the other two, which are on Twitter. I don’t want to link to Threads, but here’s a screenshot in dark mode:

      • Zagorath@aussie.zone
        link
        fedilink
        English
        arrow-up
        2
        ·
        1 month ago

        Oh thanks! I was wondering why the first one didn’t display a handle when the other two did.

        • Fmstrat@lemmy.world
          link
          fedilink
          arrow-up
          2
          ·
          1 month ago

          Are you running a bot that does this? I was considering making one for instances to run, but was concerned that people might think “AI bad” about it.

          • Zagorath@aussie.zone
            link
            fedilink
            English
            arrow-up
            2
            ·
            1 month ago

            Nope, just me, and just when I feel like it (often when I see a post I want to comment on anyway).

            Ideally, OPs would do it on their own posts. Failing that, it’d be great if there were a bunch of people who could follow my lead and do it on a post or two per day. AI can be helpful, especially for OCR, but personally I wouldn’t trust it to point out salient parts of the image in a relatively easy-to-read format.

            • Fmstrat@lemmy.world
              link
              fedilink
              arrow-up
              2
              ·
              1 month ago

              I wouldn’t have, but Qwen3 VL (fairly new open model) has been doing a fantastic job of both describing and OCR in my local AI setup thus far.