• 1 Post
  • 306 Comments
Joined 3 years ago
cake
Cake day: July 3rd, 2023

help-circle

  • A classifier will be trained differently and basically take $TEXT and map it to $CLASSIFICATION and then you allowlist on the classification. You can then also, if you love burning tokens and spending money on things, have another step that “translates” the user input, making prompt injection significantly more difficult before finally routing it to the LLM. Generally guard rails should then be placed on the LLM’s available actions via tool calls and those guard rails should be written by LLMs that are LLMing LLMs and L̸̡̡͕͓̣͓͚̝̥̜̼͙̪̖͍̺͝Ļ̵̥͍̳͇̱̠̬͔̹͖͓͖̯̹́͊̽̄̂͑͌͆͗̑̄͜Ṁ̶̡͙̺̖̥̲̙̩͖̻̹͇̜̲̃͘͜͝ș̵̖̤̫͓̦̲̼̖̗̟̯̩̺̣̞͊̆̉͐̈́̃̈́͗̐̽̽͋̊̉ͅ ̵̖͉̥͖̪̳͍̝͉͕̬̪̂̔̊̌͂̍̚͜a̶̦̤̬̱͖̻̝̱̿͋̿̍͂̒͆̈̄͑̓̕r̴̡̤̙͙͕͚̼͙̦̲̅̔̓̈́̄̆̾͗̿͌̐̄͛̒͘͜ͅe̴̡̡̼͉̣̼͉͓̪͕̬͊̐̈́̇̽̒̂͆̎̈́̃͠͠ ̶̩̯̲̯̃͛͌̀̿̅̎̾̕̕ǹ̸̬̟͍͇͍̭̹̲̯̽̒͘͝͝ȩ̶̢̬̺̦͖̣̙͍̞͇̰̉̋̓͐͂̏̀̿̑̂͌͗͗̒͘ͅͅe̶̠̝̦͛̓͛̍̑̆̽͐d̸̦̯̖̥̟̻͔̦̂̿͆͂̑̎̔̏ȅ̴̡̨̺͓͇̦̣̦͖̬̹̥͎̾̽͐̓̍̽̾͑̿͆̄̀̈́̄́̕͜d̶̡̛̝̟͔̳̲͙̰̗̬̳̯͚̐̍̑̓̉̂̍̊̓̿͝ you probably f̶̢͓̘̹̟̥̙̥͚̼̬̭͍̫̩̀͑̚͝͝͠ę̴̰̦͚̔̀́͊̋̈̒͗̆͘͝͝͝͠ȅ̸͚̘̤͚̬̩̈́̇̆ď̷̰̆̀̔̾͌̿ ̵͚̦̥̈͌͑̂͗̈́͂̕ͅm̴̢̡͇̟̠̬̣͙̹͇͉̈́̌̓́̓̌͐̓̂̚̚͜͜͝e̸̢̳̟̙̯̻͚̭͎̫̖̺̥̬͕͆̆̇̑̄̀̓̅̀͛̊͐ ̸̡̢̨̛͚͕̳͙͚͋͛̊́̍͛̽̉̃̇̇̋̑t̸̡̰͔͎̮̯̗̤̆̋͗͐̈̂̈̃̓̒̌͂̕͘͝ǫ̴͇̳͙̞̘̻̙͍̘̲̜̮͇̙̇̓̌̃k̸̢̢̤͔̫͖̲͇͍̐̃̄e̶̢̳̪͉̦͙͓̻̖̔́̃́̈́̀̿̓͆̊̅̒͐̑n̶͖̙͚̫͓͚̦̭̗̲̓́̋̀͜͝s̴̠̼̈́̑̾̑̆̽̊͋̾͛̐̊͆́͊̕̚ need to have some guard rails l̷̬̳̘̰̅̀̑͑̊̒̓̊͌̌̕e̶͖̱̊̍̉͆̐̍̌́͑͘͠t̵̨̛̹̰̳̩̹̞̹̼̪̮̄̀̅̾̔͐̇͒́̊̃̀͘̚͜͠ ̴̢̢̭͖̤̜͈͕̯͎͉̰͇̾̐͛̽̅͂̒͝m̸̨̰͖͚̞͍͍̳̭̗̳͕̈́̃̇̈́̎̌͊̍͌̊̆͠ȩ̵̡̛̜͍̞̻͎͎̲̜̹͖̼̽͂̐̽̊͌̓̀͂̐͂̾̇̚ͅͅ ̶̧̖͚͈̤͇͕̼̜̖̮̙͍͖͋̉̾͋̌̔̿͊͋̾̚l̸̡͕͇̭̤̥̩͙̫͕̆͌̈́̀̋͆̃͋͜͝͝͝͝ͅo̷̧̘͖̤͖̩̺̞̹̼̟̟͕̫̬͎̅̀ǫ̸̢̪͍̭͙̗͔̪̘͉̊̕͜ș̶̨͉̲̙̜͔̫̲̥͕͕̮̜̭̮͍̈́e̶̢̛̮̲̯͎̙͇̪̬͕͖̠̘̯̰̭̓͐̌̉̀͋̉̏̈́̇̾͠͝ ̵̬͍͔̓́i̶̘̫̯̻̘̯̗̔̒͊̂̂̑͑̍̋͑̏̀́͌̚͝͝n̵͓̬̮̻̹͉̗͎͓̖͌̿̃̓̋͂̅̍͒̉͆̿͆̉͠͝ ̴̧̲̯̲̅̀͂̆́̓̈́͑̾̾̊͋̒̿͛͘̚t̸̢͉̟̹͆̈́̏̍́̎ĥ̴̢̥̦͓͕̘͚̙̪͙̀̽́̒̓̈́̋͗̀̈́̇ȩ̵̠̹̻̯̖̹̹̗̜̻͍̹͆̊̈́͑́̏̑́̎̚͜͜͝ ̴̧̟̭͍̣͋͋̾ẘ̴̪̦̐̍̏̆̑̇͗́͆͘̕̚͝͝ͅo̵̡̟͚̲̦̤̩͎̥̘̻̥̠͙͈̙̬̓̌͊̋͐́͌̐̎̏̋̿̏r̵̢̨̘̓͑͑̆͒̂͝l̸̡̨̛͖̖̟̼̺̦̰̗̭͔̺̓͗͛̈́̈͆̈͗̐͝d̴̢̠̠́͂̾̌̿̐̀͛̄͑́͐̉̋͝ on those LLMing LLMs that are LLMing your LLM tool calls y̵̧͍̣̭͍͚̱͉͍̘̠̦̮͇̻̓͆̕o̸̥̮̮̙̖̩̙̝̘̫̗͎̺̓̊̐̄̑̎u̷̡̨̡͎͔̼̪͈̦͚͈̟͍̽̀̏͛͊ ̴̨̟͖͖̰͕͉̫̫̜̥̦̲͆͋͌̌̓̈́̉͆̉͠ẗ̵͚͉͈̦́̈́͛́͋͜i̷̧̗̰̘̖͍͖̼͕͗́̊̏̇̾̐̆͗́͌̋̀̚ņ̷̣̱̖̬̼̆͜y̷̢̡̧̛̹͙͕͔̪͛ ̶̪̫̙̘̥̹̻͙̝̜̠̗̭͎͎̆̇̅̎͜͜h̷̡̨̰̺̞̝̩͉̬͓̮͙͉͕̓̄̊̿̋́̇̏͝u̸̮̙̲̙̠͉̪͈͖͙̓̈̉̀̚͝m̶̢̛͙͕̘͈̤̬͎̺̲̯̱͉̪̠̎̆̽̅́̂͐̀̄͜ͅã̶͕̼͎͇̲̐͗̑ň̸̢̢̝̩̟͖̫͔̥͚̘̟̖̘̤̾̀͋̒̄͠͝s̸͙̳̳̤͓̝̥͓̱͔̣̹̭̐̐͋ͅ ̶̩̹̪̑͑͒̽̇̀̽̏̓̌̑̈́͂̓͘h̶͓͖̱̲̘̗̿̽̈́̎͑́̓̉̑̕̚ą̴̰̣͚̫̤̤̬̘̼̠͍͂̏̚ͅv̵̢̺̮͉̤͚̳͕̥̳̇̕̕͘͠ȩ̷̡̮͖͚̩̙͖͎̓̀͗͊͋̈́̔̐̕̚͠ ̷̫̌̀̊ň̸̜͓͕̳͙̩̼̜͈̊͜ǫ̶̝̺̰̹̻̰̰̗̪̔ ̷̲̤̪͇̠̼͑͊͘͜p̴̡̧͉̩̜͕̽̏̀̌̇̈́̊l̶͖͔̯̲͔͓̩̂̌̂͐͗̕a̵̢̨̞̰̪͖̲̭̝̪̥̻̐̃̊͗̅̓͐͋͛͘͝c̴̛̮̙̦̠̯̼̣̠̳͔̖͕͓͋̓͆͒̋̒̾͐͒̀̓͑͠e̷̢̢̧̱̣̫̲̪̣̭͙̟̔͌̽̿̐̿̿͑̆̇̇̉̊̓͘͠

    You could also do none of this and just not use an LLM, because you likely don’t need one, but then the cool kids will stop inviting you to their parties or something and you probably missed out on the blockchain parties so fomo



  • There isn’t too much more to elaborate on, it simply isn’t the case, but it is such a common thing to say that it is worth researching. You’ll hear the 1919 case “Dodge v Ford” brought up as proof, one good place to start reading is the Wikipedia “Significance” section and the associated citations, as they go deeper into the subject than just that single case.

    There simply is no interpretation of “fiduciary duty” that requires “maximization of shareholder value” outside of the misinformed lay person discussion of the subject.

    You can even just logically see why such an interpretation would be impossible: given N shareholders all with different interests and different financial situations, how do you even define “maximizing their value” and what time duration are you basing that off? Imagine how absurd the courts would be if such a statute did exist: you’d have people owning 1 share in a company suing the company because it did something that wasn’t maximizing “value” as they define it.

    The reality is that the business judgement rule is case law and is often cited and is often used as defenses against such assertions.




  • qqq@lemmy.worldtoComic Strips@lemmy.world•Beacons
    link
    fedilink
    arrow-up
    11
    ·
    12 days ago

    As a very pale person who was bullied a lot for it and probably has lasting personal doubt from that bullying, I think there is just a fundamental difference (in the US). One is “just” mean and bullying, but there is no horrific dehumanizing history behind it. Both are bad, but I don’t think they’re directly substitutable in the way you describe, as I think it cheapens the history of slavery and racism




  • qqq@lemmy.worldtoMicroblog Memes@lemmy.world•RatGPT
    link
    fedilink
    English
    arrow-up
    1
    arrow-down
    1
    ·
    17 days ago

    I understand fairly well how LLMs work, but I don’t think we understand the conditions for the emergence of intelligence or thought enough to conclude that “learning to predict what comes next” is not sufficient to create some form of intelligence, that’s all. [Edit: if we determine that that turns out to be sufficient I will be surprised] That uncertainty is sufficient to make me agnostic on the subject and that stance does have an effect on my actions, for example I’m never mean to an LLM.



  • qqq@lemmy.worldtoMicroblog Memes@lemmy.world•RatGPT
    link
    fedilink
    English
    arrow-up
    1
    ·
    17 days ago

    The American competitiveness angle is definitely… well an interesting choice; I also was taken aback by that. Ignoring some of the posturing along those lines I find the post itself interesting.

    If you think the Chinese room idea is valid we probably just fundamentally disagree and that’s OK.


  • qqq@lemmy.worldtoMicroblog Memes@lemmy.world•RatGPT
    link
    fedilink
    English
    arrow-up
    1
    ·
    17 days ago

    Odd example but yes it gets uncertain very quickly. If we discover the mechanism of our own thought it’d be easy enough to apply those rules to the new subject and see if they match or at least look similar. A negative wouldn’t be conclusive, but it’d at least give us something to work with. As it stands right now we just guess, and that guess is essentially guided by bias or at best by the fact that the thing we’re looking at doesn’t appear even remotely similar to anything we’d call “thinking”. This is simply not the case with LLMs (they do a very good job of displaying something that looks like thinking) so our guess is imo even more unsound.


  • qqq@lemmy.worldtoMicroblog Memes@lemmy.world•RatGPT
    link
    fedilink
    English
    arrow-up
    11
    arrow-down
    1
    ·
    17 days ago

    I have no idea. We are unable to prove that other humans are thinking and not just “mimicking intelligence” let alone something we’ve created. People approaching this subject with such certainty are imo being overly confident or outright dishonest.

    With respect to what I believe, since you asked, I think there is essentially no reason to distinguish “mimicking intelligence” with high fidelity from “intelligence”. I don’t feel comfortable talking to LLMs as if they were merely machines, but I don’t approach them as if they were human either; they’re in a very weird spot to me. I understand the simplicity of the math and programming behind them probably better than average, but that doesn’t help me come to a conclusion on “thinking”


  • qqq@lemmy.worldtoMicroblog Memes@lemmy.world•RatGPT
    link
    fedilink
    English
    arrow-up
    5
    arrow-down
    6
    ·
    17 days ago

    these models are only mimicking intelligence

    I always find this to be such an interesting turn of phrase: “mimicking intelligence”. If it is possible for these systems to cross the line into “thought” I think we’ll miss it entirely because of our biases that the type of “thought” we have is somehow special or mystical even.