FYI – Rule update – No LLMs/AI

r/

Evening Askers!

Following on from https://reddit.com/r/AskUK/comments/1m9cq92/should_raskuk_allow_people_to_use_ai_to_answer/ we’ve made our 10th Rule!!! I can almost feel the excitement – quoar rule updates, yeeeeeerrrrrr boiiiis…

The Rule:


No AI generated questions or answers

AskUK is a place for real answers to real questions. While we will permit the use of language cleanup and grammar adjustments via AI, anybody we suspect that is using AI to automate/generate their answers or questions will likely see their post or comment removed and be banned. It is often better to see sub-optimal text than it is generated text.

If you think your text is evidentially AI but this is appropriate, make it clear you are doing so.


Do note the minor exception at the end there – we realise people are sometimes using it for good reason. But this can be quite jarring to those of us with keener eyes, so please just be upfront about it.

Also, we have added a report reason to help people highlight the use of AI to us, we’re hoping people will use this responsibly, and not just for people they disagree with. Giving guidance like this a once over – https://en.wikipedia.org/wiki/Wikipedia:Signs_of_AI_writing. AskUK is a helpful space, please don’t just go around spouting “bot” like this is your first time outside your schools chromebook :).

Our hope is that helps us maintain a genuine human space that people find real value in, enjoy, and continue to want to participate in, keeping our community together!

Thank you all for help and feedback.

Comments

  1. JennyW93 Avatar

    Viva la revolution! Eat my shorts, clankers

  2. BoopingBurrito Avatar

    Perfectly sensible update. One thing just to be wary of – quite often people with autism get accused of being AI, because their tone mismatches the discussion, their writing comes across as overly formal/neutral, or humour/nuance has flown straight over their head.

    Doesn’t mean the rule change isn’t necessary, but more something for you as mods to keep in mind when judging posts.

  3. audigex Avatar

    Thankyou for this

    I’m so sick of “Google AI says:” being in half the posts on Reddit, as though the OP couldn’t have just done that themselves. It’s the dumbest thing ever

  4. KelpFox05 Avatar

    Good job, mods! Always remember – when somebody asks a question, they’re asking YOU, not ChatGPT or Google AI or Grok or any of the other LLMs out there. Utilise your own brain.

  5. Sparko_Marco Avatar

    Our AI overlords aren’t going to like this.

  6. joeyat Avatar

    See, an em dash! Report an M—dash!

  7. Mariashax Avatar

    Doesn’t AI get like 70% of its information from Reddit anyway? The snake is eating its own tail!

  8. Alarmed_Crazy_6620 Avatar

    I won’t delve into details but this is difficult to identify

  9. gyroda Avatar

    Good.

    If we wanted to know what ChatGPT thought we can ask it myself. We don’t need someone else parroting it here.

  10. Therealladyboneyard Avatar

    I fully support this!

  11. AF_II Avatar

    Thank fuck, though it’s not stopping the AI scrapers from scooping up all our answers and churning them out elsewhere :/

  12. Apidium Avatar

    I like this implementation. I sometimes use ai as cleanup (dyslexia means my writing can be all over the place) but I tend not to old reddit as highly enough to bother. Ai is also useful for folks with ESL or other communication hindrances.

    Really the only issue is folks posting what chatGPT told them as if eveyone else can’t ask chat gpt if that’s what they want to hear from.

  13. cryingtoelliotsmith Avatar

    oh god… I’m a writer… I use em dashes all the fucking time don’t eat me plssss 🥺🤣

  14. bacon_cake Avatar

    I absolutely agree with this, some posts are outright obviously AI (though not seen many here to be fair).

    I will say though, don’t be all militant in banning people. It’s not always possible to be certain someone’s using AI and even as someone who’s very keen on the ban, it’s not fair to ban people accidentally assumed to be AI.

  15. RubberDuckyRapidsBro Avatar

    This seems something that is hard to police?

  16. DigitalPiggie Avatar

    Excellent update. This isn’t just a new rule — it’s the dawn of a new age for this subreddit!

    Let me know if there’s anything else I can do to help!

    ^(note this post contains human-written text only)