• 3 Posts
  • 44 Comments
Joined 1 year ago
cake
Cake day: June 11th, 2023

help-circle



  • They don’t give a shit about your stolen items, and 9/10 times it’s a complete waste to even contact them over “petty theft”, regardless of how valuable the item/s are to you.

    When I was in college, my apartment got broken into. The cop that came when I called was less than helpful. Some highlights:

    • He immediately argued with me that the crowbar marks on the door frame must have been there before.
    • I told him about the TV that was stolen and he bragged that he had two TVs that were larger than that.
    • He tried to bust me for weed when he saw my hookah. I had to explain to him that it was in fact not a bong and was only used for tobacco.
    • He mentioned that if it were him robbing me, he would have stolen the microwave instead.
    • As he was leaving, he told me that they wouldn’t be able to catch whoever did it, so they weren’t going to look.






  • TBH I’m kinda over seeing AI generated content in discussions like this. It used to be novel and quite impressive, but now that I’ve seen this type of thing so many times and the reality of the negative impacts of LLMs has set in, it just doesn’t hit for me at all. It’s fun to speculate about an LLM writing TV scripts, but my worry is that this will become an actual thing in the near future: soulless generated “content” companies can sell us without having to pay hardworking and talented writers, and that shit just bums me out.

    Anyway, nothing against you, and I don’t fault you for posting it, but that’s my thoughts. Maybe I’m in the minority though, who knows?






  • The fucksmith post in question has been removed by moderators, so I wonder if Google really is pressuring Reddit to remove jokes from their platform now. This person had no idea their shit post would be used to train AI over a decade later, and they certainly weren’t violating any policies when they posted it. It’s like nobody involved in this process knew anything about Reddit.

    Pay $60 million to train on bad data.
    Implement AI trained on bad data.
    Panic when the AI returns bad answers.
    Manually remove bad data.
    Profit??

    Edit: it was removed when I checked yesterday, but it looks like they restored it.