• 0 Posts
  • 27 Comments
Joined 1 year ago
cake
Cake day: June 29th, 2024

help-circle



  • That’s a great question! Let’s go over the common factors which can typically be used to differentiate humans from AI:

    🧠 Hallucination
    Both humans and AI can have gaps in their knowledge, but a key difference between how a person and an LLM responds can be determined by paying close attention to their answers.

    If a person doesn’t know the answer to something, they will typically let you know.
    But if an AI doesn’t know the answer, they will typically fabricate false answers as they are typically programmed to always return an informational response.

    ✍️ Writing style
    People typically each have a unique writing style, which can be used to differentiate and identify them.

    For example, somebody may frequently make the same grammatical errors across all of their messages.
    Whereas an AI is based on token frequency sampling, and is therefore more likely to have correct grammar.

    ❌ Explicit material
    As an AI assistant, I am designed to provide factual information in a safe, legal, and inclusive manner. Speaking about explicit or unethical content could create an uncomfortable or uninclusive atmosphere, which would go against my guidelines.

    A human on the other hand, would be free to make remarks such as “cum on my face daddy, I want your sweet juice to fill my pores.” which would be highly inappropriate for the given context.

    🌐 Cultural differences
    People from specific cultures may be able to detect the presence of an AI based on its lack of culture-specific language.
    For example, an AI pretending to be Australian will likely draw suspicion amongst Australians, due to the lack of the word ‘cunt’ in every sentence.

    💧Instruction leaks
    If a message contains wording which indicates the sender is working under instruction or guidance, it could indicate that they are an AI.
    However, be wary of predominantly human traits like sarcasm, as it is also possible that the commenter is a human pretending to be an AI.

    🎁 Wrapping up
    While these signs alone may not be enough to determine if you are speaking with a human or an AI, they may provide valuable tools in your investigative toolkit.
    Resolving confusion by authenticating Personally Identifiable Information is another great step to ensuring the authenticity of the person you’re speaking with.

    Would you like me to draft a web form for users to submit their PII during registration?



  • If you’ve ever been in a position where you weren’t able to relicense an entire project as GPL, or were developing for a platform that doesn’t allow LGPL3 libraries to be used because users can’t replace the LGPL3-licensed binary (ios, android, game consoles, proprietary hardware), which I’m sure many people with programming careers have experienced at some point, you’ll quickly find that any copyleft-licensed library is effectively useless to you.
    I would wager that those who have had to deal with that before are much less likely to use a copyleft license for future projects.

    There’s also a lot of small projects where the developer doesn’t care about licensing. They just want the code out there, and for anyone to be able to use it, as long as they get some recognition for making it.

    Most people aren’t lawyers, and don’t care enough to read all the different licenses and compare them all. They pick the simplest one that ensures anyone can do anything with it, and they aren’t held liable for anything.

    Apache is too full of legalese for most people to bother reading. BSD has different versions which make it more complicated to pick which one you want. MIT has much less confusion about versions (there are different versions, but most people associate ‘MIT license’ with the most common one).

    And then the existing popularity helps lock in a license choice once you’ve picked a license category. “If MIT is good enough for ‘x’, it’s good enough for me.”




  • The ‘mistake’ of receiving more than one unsolicited nude?
    He was ok with the gofundme, but wasn’t ok with you sending the gofundme to others?
    And you’re supposed to report everything to him but he doesn’t do the same for you?

    Your boyfriend is free to express his insecurities. And it wouldn’t be unreasonable for him to say that if you continue messaging strangers in that context, he won’t feel comfortable continuing the relationship.

    It’s perfectly normal to set boundaries and to get upset when your partner crosses them, but it sounds like he’s getting upset about you crossing boundaries which he never made clear.

    So not only are there unreasonable one-sided expectations, you’re also supposed to be a mind reader who knows what all of those expectations and boundaries are in advance.



  • Him ending the relationship because you saw a dick pic and have had sex multiple times in the past just shows his immaturity and lack of commitment.
    You didn’t cheat on him, you got sent a nude.
    It’s not like you were planning an affair.
    I bet he looks at porn all the time, would you break up with him for it?
    You don’t need to report every single thing that you do to him, does he honestly do the same for you?

    He’s either a stupid misogynist who holds you up to higher standards than he sets for himself, or he’s just looking for an ‘out’ to the relationship without taking any responsibility.
    He’d prefer to make you feel bad than to admit he doesn’t want to continue the relationship due to his personal beliefs.

    I don’t want to tell you who you should or shouldn’t date, but there are much more understanding people out there who will love you unconditionally.




  • phlegmy@sh.itjust.workstoLinux@lemmy.ml*Permanently Deleted*
    link
    fedilink
    arrow-up
    2
    arrow-down
    1
    ·
    edit-2
    3 months ago

    Not sure if it was a plasma issue or a wayland issue, but I tried it last year and had trouble with cursor locking.

    Virtualbox had issues with the input being intermittent, and my mouse would move off the screen while gaming.

    It might be fixed now, but I don’t plan on trying it again for another few years, because what I’m using works for me.







  • phlegmy@sh.itjust.workstoTechnology@lemmy.world*Permanently Deleted*
    link
    fedilink
    English
    arrow-up
    16
    arrow-down
    1
    ·
    4 months ago

    Apple only allowed browsers on ios to use webkit, so they quite literally were holding back browser development.

    This has only recently been changed, and it appears you can only use an alternate browser engine in the EU, so they are still holding back mobile web browser development for people in most countries.