I’d be curious how it maps in a very different language to English like Arabic. I doubt we would have this immediate connection with their language, so why would a dog have it to English sounds?
Article is good but I think it’s pretty fked that they stole an unfathomable amount of developer projects to train the models that lets them build this and they don’t have the decency to release developer tools back out to the community - only paid products. Are they not the only ones refusing to work on anything Open Source currently?
Damn, brutally true. If you've seen it, you've seen it, it plays out exactly as written here. Taking it as advice doesn't put you on their level though, their gift is incredible instinct hence why they do everything here without it.
Post was good but I was more interested in his blog posts from 2002 since he was posting every month at the time when Billy Beane took the A’s to victory. I’ve read Moneyball so figured there might’ve been more of a reaction.
I’m just saying the process is exactly the same and shouldn’t be surprising. I don’t think we need to go into why one is less harming compared to another and that makes it okay to dismiss concerns.
I don’t have a strong view, but I don’t think that should baffle you, you think the reaction comes from nothing at all? One side is pushing to “wedge” (sort of) into a very low level space so that it cements itself at a foundational level and there’s no repeal or return from it - all or nothing view. The other side picks up on that and doesn’t want to let big changes go unchallenged. It’s a pretty understandable process. The niche cases allow the challenging party to present evidence for a consequence or concern that’s difficult to convey.
HN does this constantly with the surveillance, bring up one case of police abusing it, it’s exactly the same thing.
I get some enjoyment from keeping up with news but why do we get a post like this from openAI then immediately after get the same thing from Anthropic and vice versa like it’s a single entity deciding what the next topic is to relay to us simple folk. Exactly the same thing with the Huggingface incident. If this is so important to them why has it taken until September 2026 to start making a threat report like a week after OpenAI does it.
It’s pretty retarded for them to open up with “this is the world’s greatest model” publicly released? Do they know the ability of every unreleased model?
The employees doing the stealing are serious offenders here and I hope they lose all the job security they just had. There’s no way they wouldn’t know they’d be fired if Apple found out what they were doing, but the money was too irresistible to them and they thought they’d get away with it.
It was a dataset of the entirety of the public internet from the very beginning that bypassed paywalls etc, there’s virtually nothing they haven’t scraped.
So it looks like they’re creating their own App Store within the app? At least it’s kept separate from official apps. But also how is that not a security nightmare Apple won’t allow?
Well a material difference is we don’t input/output in tokens I guess. We have a concept of gaps and limits to knowledge, we have factors like ego, preservation, ambition that go into our thoughts where LLM just has raw data. Understanding the implication of a code change is having an idea of a desired structure, some idea of where you want to head to and how that meshes together. LLM has zero of any of that. Just because it can copy the output of the result of those factors I mention doesn’t mean they operate the same.
That’s convincing and impressive, but I wouldn’t say it proves it can spot deep errors. If it’s incredible at porting files and comparing against the source of truth then finding complicated issues isn’t being tested imo.
I think we’re falling into a trap of overestimating the value of incrementally directing it. The output is all coming from the same brain so what stops someone just getting lucky with a prompt and generation that one-shots the whole thing you spent time breaking down and thinking about. The code quality will be the same, and unless you’re directing it to the point where you may as well be coding the old way, the decision-making is the same too.
They’re not equal. Humans are capable of actually understanding and looking ahead at consequences of decisions made, whereas an LLM can’t. One is a review, one is mimicking the result of a hypothetical review without any of the actual reasoning. (And prompting itself in a loop is not real reasoning)
> After the initial translation, I ran multiple passes of adversarial review, asking different models to analyze the code for mistakes and bad patterns.
I feel like you just know it’s doomed. What this is saying is “I didn’t want to and cannot review the code it generated” asking models to find mistakes never works for me. It’ll find obvious patterns, a tendency towards security mistakes, but not deep logical errors.
Who doesn't think this about themselves. It's like when people say they're immune to propaganda. Isn't this thinking what makes people think their smart devices are listening to conversations rather than targeted ads you only notice after it's had the effect on you.
I adore Zed, started on the day of the windows release and never looked back. My theme is perfect, the team works super hard, and I get a nice bit of satisfaction installing their frequent updates in barely over a second.