top 93 comments

sorted by: hot top controversial new old
[–] 109 points 1 year ago (11 children)

Well, Grok is right about one thing: Charlie Kirk getting shot is a meme video. A NSFL meme video, but a meme video none the less.

  • source
  • hideshow 11 child comments
  • [–] 88 points 1 year ago (5 children)

    The very last word out of his mouth was "violence". You couldn't dream of a more ironic twist. If it was fiction, the writers would be sacked for being too on the nose.

  • source
  • parent
  • hideshow 5 child comments
  • [–] 17 points 1 year ago (2 children)

    And it was turning around the question about the number of mass shootings in the US to focus on gang violence

  • source
  • parent
  • hideshow 2 child comments
  • [–] 0 points 1 year ago (2 children)

    Does anyone have a link for the video? A coworker showed me but I'm not about to go out of my way to find it for myself, cause it kind of creeps me out to do such a thing.

  • source
  • parent
  • hideshow 2 child comments
  • [–] 52 points 1 year ago (2 children)

    Sounds like the words "You will never say anything disparaging about Charlie Kirk" are hard-coded in his training set.

  • source
  • hideshow 2 child comments
  • [–] 40 points 1 year ago

    The statistical word parrot is going to repeat what is statistically the most likely state of a famous person throughout its training data

  • source
  • [–] 39 points 1 year ago (1 child)
  • [–] 25 points 1 year ago*

    Yep. Thanks Grok. Yes, it was fake. We all know it was fake. Simple fact. No need to investigate, procecute, argue or fight over it. He's well and will keep posting shit if he wants to, or he just retired by choice. Time will tell.

    Nothing happened; carry on everyone!

  • source
  • parent
  • [–] 32 points 1 year ago
    [–] 30 points 1 year ago

    Fake: video was edited Gay: someone shot his load on his neck

  • source
  • [–] 96 points 1 year ago* (last edited 1 year ago) (25 children)

    Is Grok thinking parsing that it's an AI video and not real? Why would grok think parse that.

  • source
  • hideshow 25 child comments
  • [–] 160 points 1 year ago* (last edited 1 year ago) (8 children)

    Because it's been trained on data from when Charlie Kirk was alive.

  • source
  • parent
  • hideshow 8 child comments
  • [–] 11 points 1 year ago (6 children)

    Good point, but wouldn't it be time sensitive? Meaning, it would give more weight to recent events? I'm going with the "kirk is always right theory."

  • source
  • parent
  • hideshow 6 child comments
  • [–] 37 points 1 year ago (5 children)

    You can set up an MCP server in front of the LLM so that it can reach out to external APIs, like news sites/feeds/etc..

    But, the majority of its behavior is going to be from its original training set, which is likely 12+ months old at this point. The pipeline for this is so long to generate and refine a good new AI model, that your data sets are constantly out of date.

    An MCP server will only get you so far.

  • source
  • parent
  • hideshow 5 child comments
  • [–] 9 points 1 year ago* (4 children)

    Duckduckgo's AI search seems to slightly agree with you, meaning Futurism and Engadget:

    Grok, the AI chatbot, initially spread misinformation about Charlie Kirk's death by claiming he survived the shooting and that videos of the incident were fake. This confusion stemmed from the chatbot's inability to accurately process breaking news and conflicting information, leading to a series of incorrect statements before eventually acknowledging Kirk's death.

    That means AI really, really sucks and can be manipulated easily.

  • source
  • parent
  • hideshow 4 child comments
  • [–] 6 points 1 year ago

    Yes, LLMs, or what people call AI, are absolutely manipulated easily. Just the way you phrase your question can steer it to answer in a particular way. I haven't been on Twitter in a long time, but I hopped on yesterday and today to check out all of the Kirk memes.

    I saw so many comments from people both happy and upset with Kirk dying, that were asking questions in manipulative ways (to the llm) to try and get the response from grok they want.

    LLMs are indeed horrible for live or recent events, and more importantly horrible for anything that is super important to not get wrong.

    Don't get me wrong I personally find llms useful and I use open source models occasionally for tasks they are better at, for me typically that means reformatting or compiling shorter notes from documents. Nothing super critical.

  • source
  • parent
  • [–] 4 points 1 year ago

    I think it's well known at this point that grok in particular has been designed to be easy to manipulate by deliberately keeping it in the dark and feeding it only select information so that Musk can make it say what he wants.

  • source
  • parent
  • [–] 78 points 1 year ago (2 children)

    Grok has clearly been instructed to defend Charlie Kirk in every context and to assert he "wins" in any given situation. Right-wing people like Kirk and Musk are obsessed with image and not losing face. Grok, like Kirk was programmed to simulate the kind of rationalizations and vernacular that appeal to right-wingers, and so it tries to spin-doctor Kirk getting capped as "actually he's fine lol shut up nerd". I think it's hilarious, because it implies "Defend Charlie Kirk" is excruciatingly over-emphasized in the LLM's instruction set to compensate for the fact that Kirk actually got humiliated regularly.

  • source
  • parent
  • hideshow 2 child comments
  • [–] 40 points 1 year ago (1 child)

    And this is why you should never trust Grok, or any LLM for that matter, to be completely free from, well, just making shit up. This is because they do not think. They do not understand context. They are incapable of actual understanding. They are a next word guesser. Pure and simple. And when it has been trained to be bat shit crazy. Well you get bat shit crazy.

  • source
  • parent
  • hideshow 1 child comment
  • [–] 6 points 1 year ago

    LLMs are filtered and trained in a way that benefits their creators. But even if they weren't, which they are, they're being trained on hugely biased datasets like Reddit.

    So yeah, never trust any LLM for anything.

  • source
  • parent
  • [–] 44 points 1 year ago (6 children)

    Grok is an LLM and doesn't think at all.

  • source
  • parent
  • hideshow 6 child comments
  • [+] -16 points 1 year ago (5 children)

    Most modern chatbot LLMs have deep research that will have web search capabilities, including Grok, but probably not the Twitter bot version.

  • source
  • parent
  • hideshow 5 child comments
  • [–] 5 points 1 year ago (2 children)

    “Deep research” is just looking stuff up on the internet and passing notes to your future self that’s going to look at the results.

  • source
  • parent
  • hideshow 2 child comments
  • [–] 7 points 1 year ago

    fElon has been ripping apart grok and having its code rewritten to be more right wing. Had some new name like super mecha nazi grok or some such for it and grok was spewing right wing diatribe.

    RIP the real grok

  • source
  • parent
  • load more comments (2 replies)
    [–] 25 points 1 year ago (5 children)

    Holy crap that's awesome and scary. I'll never understand how so many take what these things say at face value. If only the public weren't so easily assuaged we could realize how dangerous this is.

  • source
  • hideshow 5 child comments
  • [–] 9 points 1 year ago (3 children)

    And if the public was more easily sausaged I'd be getting laid right now.

  • source
  • parent
  • hideshow 3 child comments
  • load more comments (1 reply)
    [–] 25 points 1 year ago (4 children)

    Imagine the world where this actually was a meme shot gone wrong. Maybe Kirk went a little nuts and decided to try to duplicate Trump's miracle near miss. The ultimate way to build credit among the right wing. Kirk hires a sharpshooter to give him a near miss, to just graze him with a bullet. And...well...the sharpshooter missed...

    Now, I don't find this scenario particularly likely, as I don't think any of these right wing social media goobers are brave enough to actually put their person in such direct risk like that. And you have to be pretty fucking dense to hire someone to shoot a gun in your direction. But it is fun to imagine that wonderfully absurd scenario.

  • source
  • hideshow 4 child comments
  • [–] 18 points 1 year ago (2 children)

    Naw, it was an assassination. Something only a government would want to pull off.

    That leaves us with three likely candidates and a couple of unlikely.

    If feel pretty strongly that because of the laser quick response this administration was aware beforehand.

    The primary candidate would be the US. We all know that they love false flag operations and with the tide turning against this administration it needs a war or some major travesty to distract.

    The second would be Israel. Their house of cards is about to fall and they have done similarly desperate terrorist acts like this before.

    The third would be Russia the master shot caller who has been responsible for a lot of the far right movements we are familiar with.

    I will admit your scenario is infinitely more entertaining and creative than what we are actually dealing with.

  • source
  • parent
  • hideshow 2 child comments
  • load more comments (-1 replies)
  • [–] 11 points 1 year ago*

    You mean he might have Smollet himself? I like it.

    There was that one YouTuber who was killed when he had his GF shoot a book he was holding for a stunt (thinking the book would stop the bullet).

  • source
  • parent
  • [–] 23 points 1 year ago (2 children)
  • load more comments (1 reply)
    [–] 21 points 1 year ago (1 child)

    Denial is part of greiving.

  • source
  • hideshow 1 child comment
  • [–] 2 points 1 year ago

    In this case, though, denial is part of it being unable to "learn" new facts without being restrained from scratch.

    Possibly with a bit of having been pre-prompted to never say anything negative about certain specific magasphere personalities on top.

  • source
  • parent
  • [–] 20 points 1 year ago

    Like trying to argue with a real nazi. Seriously why are seemingly normal people even on that garbage platform?

  • source
  • [–] 19 points 1 year ago

    Omg, the math shoggoth unable to change can't incorporate new information???

    What a shock

  • source
  • [–] 15 points 1 year ago (4 children)

    OK, it was is exact moment that I actually fully believed this is a simulation with the 'surreal' parameter maxed out.

  • source
  • hideshow 4 child comments
  • [–] 13 points 1 year ago (3 children)

    I recently learned the term for this feeling is ‘derealization.’

    And I too increasingly feel like I’m in someone’s game of The Sims and they’re just fucking with me more and more to see how insane things can get before I lose my mind.

  • source
  • parent
  • hideshow 3 child comments
  • [–] 2 points 1 year ago*

    A staple of the call of Cthulhu roleplaying game.

    A tentacled entity from beyond space and time the size of a large office building is coming out of the sea? It’s a trick of the light. They must be shooting a movie. I’m just really tired.

  • source
  • parent
  • [–] 2 points 1 year ago

    I really just want to ask how serious people are when they say this. I caught myself saying "the AI on these other drivers is getting too predictably stupid" out loud in the car today. Was my subconscious filing a bug report on the simulation and it spilled out into my conscious mind?

  • source
  • parent
  • [–] 31 points 1 year ago (4 children)

    Because violence is banned from social media, it doesn't get to recognize it when it sees it.

    This is so informative about how biased they can be.

  • source
  • hideshow 4 child comments
  • [–] 8 points 1 year ago*

    Right wingers can post CP, gore, and make calls to commit murder and rape all they want on Twitter. Go look at what they're posting now about what they're going to do in retaliation.

    It's everyone else who is getting ban hammered into submission for even contemplating fighting back.

    For anyone who thinks that political violence is a two-way street in this country, just look at the difference between what one side is allowed to getaway with and the other isn't.

  • source
  • parent
  • [–] 13 points 1 year ago (1 child)

    My man Grok settin’ the record straight for the ‘20s, and keepin’ it real af. He ain’t fooled by all that “Charlie Kirk took a round to the neck and suddenly became Old Faithful!” hoopty-jazz, so y’all can give up with that nonsense.

  • source
  • hideshow 1 child comment
  • [–] 11 points 1 year ago

    @grok is this true?

  • source
  • [–] 9 points 1 year ago

    Sounds like some shit Elon would say, it really is made in his image.

  • source
  • [–] 6 points 1 year ago (1 child)
  • [–] 6 points 1 year ago

    Now, now, now. Grok is right. Charlie "My Face is Tiny" Kirk wasn't shot. The only thing was a hoax.

    My evidence? All these "students" ran away when the "shots" happened. The average college student is going to be so desensitize to gun violence due the amount of school shootings they experience. Real students would have continued along as if nothing had happened.

  • source
  • [–] 12 points 1 year ago

    Dude do you know how hard I'd fucking laugh if he turned up like "Ahh what's up online, let's see... What? Shot? No me and the family were in The Poconos for the weekend, I'm fine!" And it was an AI deepfake the whole fucking time and now NOBODY CAN TRUST ANYTHING FOREVER lmao. Not that I like the guy but like, one day that'll happen, an AI deepfake of something will sweep the internet as "real" before being proven false and everyone who believed will be deeply unsettled, and here's the hopeful bit: maybe then we can convince a critical mass of people against AI.

    But yeah if he did I'd laugh so hard my neck would just do that.

  • source
  • [–] 4 points 1 year ago (1 child)

    Poor mechahitler doesn't want to accept that his friend is gone 😞

  • source
  • hideshow 1 child comment
  • [–] 4 points 1 year ago

    Grok is my uncle at the dinner table??

  • source
  • [–] 5 points 1 year ago* (2 children)
  • [–] 19 points 1 year ago

    I think that's just squishy back-fat, folding over the back of the chair.

    What looks really suspicious is his face - notice how his facial features are about twice the size that they appear normally, when you see them in memes.

  • source
  • parent
  • We'll that interesting, even racist AI grok can't cope right.

  • source
  • [–] 2 points 1 year ago

    See? Conservatives getting worked up over fake news. Release the Epstein Files

  • source
  • tomato tomato

  • source
  • [+] -23 points 1 year ago (3 children)

    How about we see the prompts that preceded this? No? Didn't think so.

    Lemmy: AI is so dumb! You can make it say anything!

    Also lemmy: I believe every AI post I see is unprompted and spewed straight from the AI!

  • source
  • hideshow 3 child comments
  • 7 comments in c/ArtificialIgnorance@lemmy.ca view that post