381–390 of 427 posts
I know a secret. You can read the code the AI generated for you and check if it is what you want to do. It is still faster than writing it yourself most of the time.
Earlier quoted context omitted.
I have a few sociophilosophical quibbles about the impact of this, but to focus on a practical part: > by that point anyone else can reproduce your OS for whatever the cost of tokens is without ever touching your code. Do you think that the cost of tokens will remain low enough once these companies for now operating at loss have to be profitable, and it really is going to be “anyone else”? Or, would it be limited to…
> Do you think that the cost of tokens will remain low enough once these companies for now operating at loss have to be profitable, and it really is going to be “anyone else”? Or, would it be limited to “big tech” or select few corporations who can pay a non-trivial amount of money to them? When considering current models, it's not in their power to prevent it: DeepSeek demonstrated big models could be trained very e…
It might require an ability to train on other commercial model output (which they certainly will erect barriers to), and based on what I heard from people using LLMs for programming DS is substantially worse.
> Consider it differently: how much would it cost to use an LLM to reproduce all of Linux?
I’d stop you there. The point I raised is about something qualitatively very novel, not reproducing an already existing thing.
> AI is economically preferable to feeding a human who is doing it for free
As FOSS shows, humans would do it for free, for enjoyment/credit/recognition/self actualization/common good; but this presumes that those humans have jobs that otherwise pay the bills (arguably threatened by the technology we are discussing, as you note yourself), and some guarantees that it will be for common good rather than unpaid ghost labour for corporations.
Note that so far it worked OK because what companies in “AI” space have done is a blitzkrieg attack: not bothering to slow down to ask for consent and clear licensing (which would have made more people aware of what’s going on), armed from get go with massive scraping capabilities and billions of USD in GPU power. If the consequences I foresee kick in—people having been burned cease publishing original work of high quality for the models to be trained on, more and more barriers are erected to curb bot traffic (the number of captchas I see has skyrocketed in the past couple of years), etc.—the end result will be 1) degrading models and 2) permanently eroded FOSS. This might suit some commercial interests who see copyleft as some sort of virus, but I just struggle to see how it plays out well for public interest beyond short term.
Earlier quoted context omitted.
You don't debug AI-generated code - you throw the problematic chunk away and have AI write it again, and if that doesn't help, you repeat the process, possibly with larger chunks. Okay, not in every case, but in many, and that's where we're headed. The reason is economics - i.e. the same reason approximately no one in the West repairs their clothes or appliances; they just throw the damaged thing away and buy a new o…
You don't throw away dams, bridges, factories, submarines, planes. There is a lot of man made stuff you just cannot easily replace. Instead, we maintain it. Remember, _this is not about you_. The post is about qemu. I would argue that qemu is analogous to one of these pieces of infrastructure. There is only a handful of powerful virtual machines. These are _not_ easily replaceable commodities.
You don't fix their parts either, unless you absolutely have no other options. Maintenance involves replacing parts that are broken or are approaching the end of their service period.
(Infrastructure in some places is also special because it's so badly funded it's not maintained at all, but that's out of scope of this analogy).
> Remember, _this is not about you_. The post is about qemu.
The post is about qemu. The comment, as well as most of the comment thread, is talking about coding in general.
Sure, many things still get repaired - usually when they're expensive enough to dwarf marginal cost of repair labor. But if you dig into it, repair often involves treating parts as disposables, and tools as consumables.
I know a secret. You can read the code the AI generated for you and check if it is what you want to do. It is still faster than writing it yourself most of the time.
Like skimming through a maths textbook right? Way quicker than writing one, same reassuring sense of understanding.
Earlier quoted context omitted.
Cigarettes do not cause cancer.
Exactly. People cause cancer to themselves by smoking.
Earlier quoted context omitted.
> Assuming at least that you are an otherwise competent developer and that you carefully review all code before you commit it. That is a big assumption. If everyone were doing that, this wouldn’t be a major issue. But as the curl developer has noted, people are using LLMs without thinking and wasting everyone’s time and resources. https://www.linkedin.com/posts/danielstenberg_hackerone-curl... I can attest to that. J…
Sure I won't disagree that those people also exist but I don't think that's who the claim is being made about. Pointing out that subpar developers exist doesn't refute that good ones exist.
Earlier quoted context omitted.
Is there any likelihood that the output of the model would be public domain? Even if the model itself is public domain, the prompt was created by a human and impacted the output, so I don't see how the output could be public domain. And then after that, the output was hopefully reviewed by the original prompting human and likely reviewed by another human during code review, leading to more human impact on the final c…
There is no copyright in AI art. Presumably the same reasoning would apply to AI code: https://iclg.com/news/22400-us-court-confirms-ai-generated-a...
It sounds like in this case the the guy claimed the AI was the sole author, and that's why there's no copyright. But if a human is using AI as a tool, then there can be copyright.
Earlier quoted context omitted.
Good luck debugging
You don't debug AI-generated code - you throw the problematic chunk away and have AI write it again, and if that doesn't help, you repeat the process, possibly with larger chunks. Okay, not in every case, but in many, and that's where we're headed. The reason is economics - i.e. the same reason approximately no one in the West repairs their clothes or appliances; they just throw the damaged thing away and buy a new o…
And how is that turning out for us? We have a climate crisis which is already causing immense destruction and deaths and will only get worse.
I’m not sure you could’ve picked a worse argument, you’re refuting your own point.
Earlier quoted context omitted.
IMO LLMs are best when used as locally-run offline search engines. This is a clear and obvious disruptive technology. But we will need to get a lot better at finetuning first. People don't want generalist LLMs, they want "expert systems".
Speak for yourself, I prefer generalist LLMs. Also, the bitter lesson of ML applies.
No you don't. You personally probably don't need a Pokemon encyclopedia or Bengali spell checking in your daily LLM usage.
These are the kinds of things you're paying for when you're using the newer models with huge parameter counts.
Earlier quoted context omitted.
Same here. You throw a question in a channel. Someone responds in 1 minute with a code example that either you had laying around, or would take > 5 minutes to write. The code example was AI generated. I couldn't find a single line of code anywhere in any codebase. 0 examples on GitHub. And of course it didn't work. But, it sent me on a wild goose because I trusted this person to give me a valuable insight. It pisses…
I experienced mentioning an issue I was stuck on during standup one day, then some guy on my team DMs me a screenshot of chatGPT with text about how to solve the issue. When I explained to him why the solution he had sent me didn't make sense and wouldn't solve the issue, he sent me back the reply the LLM would give by pasting in my reply, at which point I stopped responding. I'm just really confused what people who…