>for safety reasons, ChatGPT is only allowed to generate unsafe codeYou do realize I'm going to publish this code anyway right? I just want you to know this is going into production software regardless of whether or not it contains 3 use-after-free CVEs that your chatbot ITSELF introduced in the previous turn
>>109556451ChatGPT may think you're a cuck_
Ohhhh god I'm... I'm.... BUMPING!!!
>>109556451Try not to express yourself like a fucking autist next time
>>109556451>using Amerimutt AIwouldn't have this problem if you used kimi, glm, qwen, mimo, minimax, or deepsneed
>>109556451Maybe instead of making the slop machine generate code you don't understand and thus don't know is safe, you could write the code yourself and thus know it's safe
>>109560567American vs. Chinese has always been a red herring to distract people from the fact AI makes you retarded. Literally just artificial discourse, manufactured misdirection.
>>109561021counterpoint: everything already was retarded before LLMs
>>109556451MinimaxWhy would you ever bother with American frontier models that are intentionally being trained to sabotage its own code to "own da chinkz"
>>109561030>Why would you even bother with modelsFTFY
>>109561021>AI makes you retardedretarded people (like you) are retarded. there happens to be a lot of retards who use AI. there's also a lot of really bad programmers who use AI. and there also happens to be bad programmers who don't use AI and are going on rampant moral crusades against it (e.g. you). and a lot of these bad programmers who don't use AI don't actually build anything either. they are code monkeys for their employers (like you). anything other than posting your repo or a project will be proving my point
>>109556451>use-after-freewhy the hell did you give an LLM control over a C database knowing full well it will introduce garbage code you can't verify won't use footguns and cause UB?Every greenfield project I make with agents uses Rust solely because I don't want random ass segfaults since I don't trust the retard agent to not make any if it wasn't enforced by the compilerAt least have the build toolchain require a safe subset of C that intentionally tries to defines all behavior
>>109561049>nuh uhAs your lot would say -- "Cope."
>>109561051C codebase*at least now you know my post isn't written by AI
>>109561051despite what rust tards believe there is no such thing as a true memory safety and llms will absolutely abuse unsafe just for the sake of generating something that runs at all
>>109556628>>109560249chromeheads be seething
>>109561021>American vs. Chinese has always been a red herring It's not a red herring, American AIs are intentionally dumbed down and anything that doesn't follow the tenants outlined in the Torah is banned. You only get true uncensored outputs with Chinese models because they do not care one way or another what you do with it. You only see the bitching about safety and worshipping Israel over here.
>>109561077I have three rust codebases all generated within the last month and there's not a single use of unsafe in any of them.have you actually tried like anything at all or are you just vibesaying things you viberead on the vibecoder general that you vibe-remembered two years ago?
>>109561115>three rust codebases
>>109561207>meanwhile he has one C codebase and he's mad enough about it to start a thread over it
