Avid PC gamer, Linux convert, SCP fan.
Love Science Fiction, Cyberpunk, post-apocalyptic settings; Fan of the games of the defunct Arkane Studios. Listening to (Power-, Speed-, Thrash-)Metal, Gothic, Deathrock, EBM, Vaporwave, Lo-Fi; Classic and Musicals are fine too. Can’t stand Hip-Hop.
Owned by two cats, recently divorced, blessed with a personality disorder (AVPD) - pensioned (even the state has the opinion I’m a total wreck lol). This causes me to be unable to keep up personal connections and makes me ghost literally everyone, so if it happens to you, sorry in advance.
Chronically online.
Pro GenAI, but Anti-GenAI-Corpos; this technology should be available to everyone, which would only be fair since we all contributed to it. Datasets and Models should be under the jurisdiction of UNESCO, since they are literally the distilled cultural output of humanity.
Everything about authentication and Microsoft is nightmare fuel. I am sure that noone over there has any idea what is going on, and that is the reason why every new thing from microsoft builds their own - soon to be legacy code anyway - authentication portal and ties it to the existing kowloon walled city of authentication with chewing gum and zip ties.
I remember me being 14 and distributing stacks of 3,5" 1,44Mb floppies filled with copies of Doom to whoever wanted one, as long as i got the disks back later on
Me: Plex share because i am lazy as fuck regarding video content and need my limited storage for my game trove.
Not the same lyrics, but in the same vein: https://www.youtube.com/watch?v=mQnF1BBBtXs
I mistyped. Its a collection of games i have, not a game i write. I am not good enough of a coder to write a game. Which is what i meant to take responsibility for the quality of code you submit. I know that i could not guarantee that code i create - regardless of creation method - is up to spec even for something where you can get away with some substandard code.
Edit: Lol, you weren’t able to read my sentence coherently. I just read your reply and thought i had forgotten the “s” at the end of “games”.
“my game collection” isn’t “my game”. you should crusade against functional analphabetism instead of LLMs.
And to provide a glimpse of the scale of the task i use it for: I have a collection of approximately 36000 games (not counting the games in my steam / gog / itch /whatever accounts, because those already have descriptions). Writing short blurbs for this amount of games (grabbing metadata can be scripted) is not something i want to do by hand, because it would take me ages. My local model is doing this overnight, putting the information into markdown tables i then check in the morning for correctness. I run this workflow around once a month to update the tables with the new arrivals.
I am not a “slopfan”. I do not use any of the techbros services, and never have outside of a few test runs in 2022 to see what it can do.
I only use local models running on my own hardware, and only to automate tasks that would take ages by hand (e.g. creating summaries about gameplay and features of my game collection from the game stores websites, helping me with RegEx or quickly creating one-shot python scripts for specific purposes - i am a terrible coder), or that make interacting with large technical documents easier (e.g. mainboard manuals). I have dabbled a bit with image generation, but it isn’t really interesting for me.
But denying that the technology can reduce workload on coders even when not generating code (e.g. unit tests, red teaming, reverse engineering) is immature. These tasks are also much less prone to cause AI psychosis, because i do not care about “chatting” with these models.
Reported for the personal insult btw, this is not needed in any way or form.
Yes, they are disclosing it, and that’s a good thing for you! If you want to avoid products that use LLMs in their code, being informed is key. Increasing pressure on coders who are also under pressure to provide a service - especially if it is fucking unpaid like most FOSS projects - will only mean that they don’t disclose it anymore, because if they can increase their output (for instance by running unit tests or scan for issues) in the limited time they can provide, they will take that edge, people like you be damned. It’s only human.
Your way will only lead into a future where most coders use LLMs, and you are in the dark about it, leaving you without the option to choose.
Such a fucking bunch of bullshit. My fucking microwave uses more power when using my rice cooker than my GPU burns throughout a day using a chatbot. Get your numbers straight, or get ridiculed.
It isn’t possible. If you get society at a level of repressiveness that you want to have (you know who else only thinks in black and white terms?) they will simply not disclose it, leaving you with nothing to go on besides witch hunts like in the dark ages.
All the costs you name are attached to OpenAI/Anthropic/Nvidia/Microsoft, not the technology. You can run local models, and noone except Sam Altman and Dario Amodei need that amount of datacenters, because if they stop building, these guys are finished. Instead of bashing the technology which has a lot of uses that don’t need the power of a small city, go bash the fucking end times capitalists in the US which are responsible for that shit.
Yep, and even more important are the cases where coding models are used outside of actually producing code. Unit tests, red teaming, bug hunting and so on are all areas where LLMs can reduce the load on an experienced coder, allowing him to concentrate on the meaty stuff. Blindly being against the technology itself because big tech currently consists of a bunch of asshats is not a mature stance.
He’s just admitting the usage of LLMs. They can be powerful tools in the right hands, and he wouldn’t submit that code if he couldn’t take responsibility for it. The bullshit OpenAI ,Anthropic et al are pulling is not synonymous with the technology itself, which will survive those asshats.
I’m pretty sure there were people like you thinking about IDEs the same way.
Or using something else than vim for editing files.
Or replacing punch cards.
Do you really think that Torvalds would submit code that is not up to spec? He’s the most anal person regarding quality of code i know of - removing him from a project would not be a positive thing. He actually embodies the mentality of taking responsibility for code you provide; I am pretty sure he would stop coding for OSS projects before betraying that concept. We need more coders like him that take pride and responsibility for their code, not put up artifical barriers because of the tools someone uses.
Their stance regarding LLM code is simply not sustainable, because in the foreseeable future discerning between human written code and LLM code will become impossible. Every tool that can identify LLM code is at the same time the tool used for adversarial training.
In the end, it is not important where the code came from. What counts is that people who submit code take responsibility for their code, regardless of how it was created - because you won’t be able to filter for LLM code, but you can filter for quality.
They alienate people who can’t cope with change - but change is needed if Firefox doesn’t want to disappear into irrelevance. They need to have as many communication channels open as they can, and they need to implement features that might pull in new users. They actually go out of their way to make sure new features are optional, so that legacy users can keep their experience, but even implementing the possibility of using LLMs in the browser (which is pretty much a standard thing in browsers nowadays) made these FOSS users screech like harpies - it’s not like implementing a side view window (they already had those) that integraties a few API calls was a huge investment in manpower or resources. This behavior extends to the vast majority of FF forks.
I was using Librewolf, Zen Browser and Waterfox (on Android) before. Their immature screeching made me switch back to Firefox after many years. Privacy-wise i am pretty sure the loss of Librewolf’s features are more than compensated by being part of a much larger group of people with my browser / extensions choice.
🎶It’s fun to violate the D M C A🎶
🎶It’s fun to violate the D M C A🎶
🎶They want everyone, to not own their own stuff🎶
🎶You can just pirate it in that case🎶
🎶It’s fun to violate the D M C A🎶
🎶It’s fun to violate the D M C A🎶
🎶You get games to play, and movies to view🎶
🎶You can do what ever you feel🎶
If you are really using those things for work, a 32GB VRAM Blackwell currently costs around 3800-4000€. My own experience with local models on my 5070/12GB with 64GB DDR4 has shown me:
If this scales linear, it would mean that the 32 GB of a smallish Blackwell would be enough to run Ternary models with around 70B parameters plus a ton of context window without breaking a sweat.
These are strange times. I absolutely love the technology and the advances, and i fucking hate the companies pusting these technologies.
Altman, Amodei, Musk, Nadella, Huang and all of the othe AI evangelists can go choke on a bag of dicks in my opinion. This technology never should have been this divisive, it should have just become another tool in humanities toolbox. I fear that these people have tainted the general populations outlook on this tech for a very long time.
This isn’t a case where something can’t be enforced perfectly, it is a case of something that can’t be enforced at all going into the future. Every tool that can reliably detect LLM code is at the same time the tool used for adversarial training, making the generated code look more and more human. We are already at the point where for plain english the false positives and false negatives go through the roof, making these tools very unreliable and when applied automatically a liability. Code is a lot more formalized, with a lot less personal variance (spelling, vocabulary and grammar are basically fixed - only the used logic and how it is implemented is variable), making detection harder by default than in natural languages.
If LLM code can’t be detected anymore by automated means - and that state of things is approaching fast - then any policy about allowing or restricting LLM code is not worth the paper you would use to print it out. But that’s not so much of a problem. The more important policy to set, that can also be enforced, is that everyone submitting code has to take personal responsibility regarding the quality of the submission. Delivering bad code - when not happening while training to become a better coder and looking for feedback - has to lead to consequences based on the seriousness of the case and if it’s a repeat offender. Anyone using an LLM to spit out bullshit LQ code will run into that kind of rule very fast.
And that’s the only thing that is actually relevant. Using AI tools does not absolve of any responsibility, regardless of area. Speaking of which, why isn’t Huggingface suing OpenAI into oblivion? I’ve slowly come to the conclusion that this “hack” was a PR-stunt to allow OpenAI to pause training without having to admit that they had to stop anyways because they run out of cash.
This is why recovery info for archives is a boon - sadly it’s pretty out of fashion.