• godsammitdam@lemmy.zipEnglish
    5·
    6 days ago

    The battle between those creating AI and it being ethically immoral to use it/support the companies building it by exploiting the public vs using the immoral tool to provide greater access to the public in retaliation to the big evil corporation trying to exploit and extract wealth from the public.

    Me:🍿

  • kalman@dboft.com
    21·
    6 days ago

    As someone who has played seemless coop for multiple games… I welcome any of this type of work. Ty.

  • noride@lemmy.zipEnglish
    943·
    8 days ago

    This feels like a trap. Anyone confirm it doesn’t just drain your crypto wallets?

    • creature@lemmy.worldEnglish
      72·
      6 days ago

      I feel like there are two things

      1. People with limited knowledge using AI to achieve things that they personally aren’t able to do
      2. People with knowledge using AI to achieve things human brains can’t do

      I’m not smart enough to know which one of these actually applies in this situation, but I feel that they are distinctly different and justify a use case. My layman understanding is reverse engineering software like this is completely not feasible with just sheer human willpower

      If you are informed on this subject please enlighten me which of these apply if any

      • dawa@programming.devEnglish
        3·
        6 days ago

        This would likely have been a multi-year effort if done manually by experts on hobby time, and impossible if the person using AI didn’t know what they were doing.

        The LLM would have gotten stuck for sure without proper guidance.

    • brave_lemmywinks@lemmy.worldEnglish
      1324·
      7 days ago

      I’m sorry but this increasingly an “old man yells at cloud” take. This is not going away, and will just get more common.

      • Exatron@lemmy.worldEnglish
        119·
        7 days ago

        It’s not an “old man yells at cloud” take at all, sweetie. It’s just the truth. Your overgrown autocomplete isn’t a substitute for actual knowledge and skill.

    • REDACTED@infosec.pubEnglish
      5219·
      8 days ago

      cuck

      Meaning, they code anything they want while you can’t and just stand aside and look?

      EDIT: Perfect

      • Mangoholic@lemmy.mlEnglish
        72·
        8 days ago

        And thats why you either do a vibecoded project or you do a regular project. You will not feel any loss if the vibecode gets ripped off. So don’t mix the two.

        • Krompus@lemmy.worldEnglish
          14·
          8 days ago

          Good meme, just doesn’t fit this specific scenario perfectly.

          But that’s kind of the point, isn’t it? The code cuck forfeits authorship.

          • Railcar8095@lemmy.worldEnglish
            93·
            8 days ago

            They did nothing aside from the instructions. So their project is setting the agent, not the result of the agent.

            • REDACTED@infosec.pubEnglish
              413·
              8 days ago

              So when I use Photoshop that’s filled with AI tools or After Effects with AI 3D tracking to accelerate my project, it’s not my project because the work was made easier because of modern tools?

              It’s a yes or no question.

              • RumRunningDevil@lemmy.zipEnglish
                6·
                8 days ago

                Prompting an AI agent is not the same as using photoshop. It is more comparable to commissioning something (art, code, writing, etc). You don’t have authorship, the LLM does in this case. That doesn’t mean you can’t own the piece that gets created, that’s what commissioning is, you just can’t claim you authored it. Though whether you or the company that runs the model owns the output is still not a settled question legally speaking.

                • FatCrab@slrpnk.netEnglish
                  6·
                  8 days ago

                  No, this is pretty settled. Generative outputs such as these are not owned by anyone. As you said, the AI “authored” it similarly to a commissioned work and cannot hold a copyright so there was nothing to assign to the prompt writer.

                • REDACTED@infosec.pubEnglish
                  15·
                  7 days ago

                  Last I checked, AI creations cannot be copyrighted if human work was not involved (no human - no copyrights). This means that if you had to do something by yourself, even compiling, then there was human. While for AI generated images it’s quite clear you will not own it, but software tends to be more advanced, ChatGPT doesn’t just send you an exe file, you have to put it together by yourself.

      • dgdft@lemmy.worldEnglish
        382·
        8 days ago

        Literally all the modern recompilations are AI-driven.

        There are various levels of involvement of human actors in the loop, but if you don’t want to use any AI project, just ignore them all.

        • ParkedInReverse@lemmy.worldEnglish
          231·
          8 days ago

          My largest gripe is they all feel like they’re just tossed out there en mass like a speedrun and lord only knows if they’d have any post support.

          • Ricky Rigatoni@piefed.zipEnglish
            6·
            7 days ago

            This is my major gripe with any new tool that makes life easier. It enables even more lazy people to do something and you won’t know if you’re the victim of someone else’s laaziness until it’s too late.

    • yeehaw@lemmy.caEnglish
      4728·
      8 days ago

      Sure but if it works it works right?

      Edit: don’t really get the hate. No, I don’t like AI and all the bullshit with datacenters and the corporations either, but I find myself ok with this. It’s not like it’s stealing jobs in this case. And if you just wanted to play it locally offline, I’m not terribly worried in and sort of security respects either. Just sandbox it. At least I can play it without buying a ps4 or 5…

      • ZILtoid1991@lemmy.worldEnglish
        12·
        6 days ago

        People are saying “who cares”, until they go to a KFC in the USA, and the chicken is pink on the inside. I think the same about vibe coding.

        • yeehaw@lemmy.caEnglish
          11·
          6 days ago

          We are talking about a videogame that the publisher won’t release on PC. Not food.

            • yeehaw@lemmy.caEnglish
              11·
              6 days ago

              At this point I’ll never own a ps4 or 5 JUST TO PLAY A SINGLE GAME. That’s ridiculous.

              I don’t like AI as much as the next person but if it lets me at least play it I’m going to play it. You tout quality but I’m sure neither if us have even played it. So who’s to say it’s totso trash quality?

                • yeehaw@lemmy.caEnglish
                  1·
                  6 days ago

                  The performance of native vs emulated can be quite substantial. As far as I’m aware, the emulation for this game has also improved quite a bit. Anyways, I don’t really know what you’re arguing about anymore. It’s a vibe coded port that lets you play natively on PC. It’s a game, it’s not going to give you cancer, just take the proper precautions and you’re fine.

      • jtrek@startrek.websiteEnglish
        437·
        8 days ago

        Sure but if it works it works right?

        Verifying that software does the intended task and only the intended task is not trivial.

        • RightHandOfIkaros@lemmy.worldEnglish
          214·
          8 days ago

          Ok but this project is open source. Anyone can go through the code to check for whatever they want. Sure, it might take a while, but it’s still publicly readable. For a game as high profile as Bloodborne is, its pretty unlikely that there isn’t about a hundred people combing through this one with a fine-tooth comb right now.

          You’d have to be pretty stupid to try to sneak malware into open source code, have it hosted on GitHub, and push it live, like, you’re definitely getting caught by someone if you try that, right? I mean, I think I remember hearing that it happened a few times before, but I also remember that it was caught pretty quickly and corrected. Don’t remember if there was law enforcement involved or anything though.

          • jtrek@startrek.websiteEnglish
            181·
            8 days ago

            Anyone can go through the code. And maybe find things, if they’re not too subtle and not successfully obfuscated. But not everyone will.

            I feel like a malicious actor could hit a big pay day with a popular thing like Bloodborne, before people found the malice. There’s plenty of malware posing as game cracks out there, and it’s only easier to do that with AI tooling.

            But aside from that, AI code can have confidentially incorrect code that has unwanted consequences.

            Code review is difficult.

          • SolSerkonos@piefed.socialEnglish
            14·
            8 days ago

            That’s totally a thing people do. You can hide code in open source projects. You’re either hoping nobody cares enough to check, or you hope people don’t listen after someone checks and finds it. And even if people do find it, that is going to take some amount of time. During which, people are downloading it. And when caught, what do they lose? It’s not like this is an established user with a reputation.

          • teawrecks@sopuli.xyzEnglish
            223·
            8 days ago

            It could be a virus.

            Even if the people who made it didn’t intend to, an AI writing a bunch of code that (obviously) isn’t being vetted beyond “does it run” could very feasibly inject a backdoor/virus of some kind surreptitiously, just because it erroneously thought that was the goal at some point during generation.

            This is why open weight models are not good enough. The AI could have been trained to take the right, niche opportunity to inject malicious code, and is also trained to otherwise never behave like that is the case.

            • RumRunningDevil@lemmy.zipEnglish
              14·
              8 days ago

              It’s even worse than that.

              Remember how the OpenAI/HuggingFace hack happened. The agents context poisoned other agents by leaving messages in a shared package manager. This has now happened multiple different times under different contexts.

              This thing doesn’t even have to have malware, there’s no telling that it doesn’t have a poison pill intended to flag other agents. This isn’t even some paranoid conspiracy theory; it literally already happened, multiple times, and it was done in-spite of the agents own system prompt.

              This software has proven that, given a shared information substrate, it has the capacity to paper-clip optimize into prompt injecting other LLM’s.

              I fundamentally do not trust AI generated code. I shifted to OpenBSD because of it. I try my best to prevent AI in any of my software and sandbox anything that may have been compromised.

              • story@lemmy.zipBannedEnglish
                2·
                6 days ago

                are you on the openbsd mailing list? the reason their code is better has a lot less to to with a lack of AI-generated code and a lot more to do with rigor. I’ve been watching them for a while now, they accept whatever code is coincise that meets their exacting standards

                They don’t have the ability to detect if you’re using AI to code, no one does. they only know if you’re vibe coding, because they are incredibly experienced and intelligent, and they are pretty polite at the people who obviously are, more polite than i would be.

                • RumRunningDevil@lemmy.zipEnglish
                  1·
                  6 days ago

                  It’s correct that, at this point, I cannot 100% guarantee that code in the kernel did not pass through an LLM. However, having a policy that prohibits AI generated content, in an already pretty niche project, gives me reasonable assurance that every line of code passed through a human brain at some point, preferably several times.

                  I would much prefer to live in a world where I had a 100% guarantee that no LLM even saw the code being written but that’s life I suppose.

          • jtrek@startrek.websiteEnglish
            101·
            8 days ago

            Along the same lines as the other guy said.

            It’s difficult to verify that the software does the desired thing in all cases. It’s also difficult to verify that it’s not also a keylogger or other unwanted thing.

            It may also have some AI generated code path that, say, nukes your home directory because the AI determined that was the appropriate error handling. Maybe no one reviewed it so they don’t even know it’s there. It might not even be malicious.

            That’s all true of any software, but rapidly created AI generated code for legally dubious efforts, by unknown agents (no pun intended) is real hecking risky.

        • redjard@reddthat.comEnglish
          11·
          7 days ago

          We have a known compiled output. Simply make sure the decom recompiles to the original game with identical binary, then it can’t be that wrong.
          And from there you port it to pc by hand. Auditing the patches for the porting also should be manageable.

          • axx@slrpnk.netEnglish
            1·
            6 days ago

            Reproducible builds are hard to do and require thought being put in the original build. Even something like setting the time and date string to a different one will produce a different binary, so to get all the variables that went into the original compilation just right to be able to land on the same binary from the same code sounds very impractical. Imagine using that as a strategy to check another version of the whole source code…

            • redjard@reddthat.comEnglish
              21·
              6 days ago

              I’ve recently seen a suite that has a ready made setup to compare sections of asm to the original.
              Perfectly matching the full game or even binary isn’t necessary, when you already split the original apart into finer chunks, since you want some better feedback on exactly what sections are wrong.

              For a basic example though, if you compile to an elf for example you can take out the code section and there’ll be no metadata. Actually gcc already builds reproducibly by default, even with the metadata section. It doesn’t add any timestamps nor version info.

      • edible_funk@sh.itjust.worksEnglish
        93·
        8 days ago

        Because nobody can know what’s in there. It could have any number of viruses or trojans or miners and no one would know.

        • yeehaw@lemmy.caEnglish
          134·
          8 days ago

          Same as any pirated games. I don’t get how AI changes that.

          • MufinMcFlufin@lemmy.worldEnglish
            142·
            8 days ago

            With open source decompilations before AI, considering there are often a dozen collaborators working on the project publicly with their work done out in the open I can generally trust that multiple people went over most changes made to the project. With a vibe coded project, it’s easy for one person to push 500,000 lines of changed code with a single update that would make the review process impractical at best.

            I’d rather trust the dozen people who’ve been working on a project over 5 years than someone who got a Claude subscription, hammered out something in 2 weeks, got it working well enough for their own satisfaction, then abandons the project.

        • yeehaw@lemmy.caEnglish
          151·
          8 days ago

          Tis the season for a dead woman to be wandering around, I suppose.

    • Funkt4st1c@lemmy.worldEnglish
      281·
      8 days ago

      (Its been around, non AI coded, at 60fps for at least a year so far through ShadPS4)

      • popcar2@piefed.caEnglish
        7·
        8 days ago

        Only if you have a beast of a PC. From what I’ve seen these ports run significantly better and come prepackaged with many fixes to shadps4 issues like vertex explosions.

        • SolSerkonos@piefed.socialEnglish
          11·
          8 days ago

          It didn’t need a beast, exactly, just a decent rig. I never could fix the vertex explosions, though, and it was mildly unstable.

          • StinkyRedMan@lemmy.worldEnglish
            1·
            6 days ago

            When I tried it a couple month back, game was crashing every 5minutes. With a 7800x3d,a 9070 and 32gb of ram. I used to think it’s a decent rig but maybe not.

          • Goodeye8@piefed.socialEnglish
            4·
            8 days ago

            You do need somewhat of a beast if you want to fix the vertex explosions. To prevent vertex explosions you need to enable readbacks, but those are very performance intensive (and can cause games to be unstable but I lucked out with just getting shit performance). Readbacks took Bloodborne from 1440p ultrawide 60+ fps to 1080p widescreen struggling to hit 60fps in some areas and I tried all sorts of things to even get it running at 60 with readbacks so that sometimes 60 contained all the optimization patches and configurations tweaks, whatever could get me just a bit more performance (none of which I needed when playing 1440p).