LordDragonfang

> That's one I have not heard before

No offense, but if that's the case, you are very new to the discussion. It's been pretty well-documented that opt-out provides orders of magnitude more useful reports than opt-in.

For the best example: Factorio, a game with an almost-exclusively-technical playerbase and extremely well-regarded and community-friendly dev team, which already had a ton of people writing good bug reports on the forums, [fixed 12 crash-causing bugs](https://factorio.com/blog/post/fff-231) within two days after making crash reports automatic and opt-in.

And if it has that much impact for Factorio, you can imagine how much bigger the impact is for non-technical software.

"Most people" with EVs charge while they sleep, because right now it doesn't make as much sense to buy an EV if you're in the actual majority that does not have access to a garage you can install a charger in. That fact is one of the major things slowing EV adoption.

> the rule of law does not strictly exist in China, the state can compel surveillance cooperation regardless of what might be written down

While I agree that China is obviously worse in this regard, it's naive to claim this is unique to China, when literally a couple of months ago the US got into a fight with Anthropic about them not removing safeguards which were already just enforcing the letter of the law.

Everyone else is already pointing out how competent over unique is purely a positive, so I want to criticize the other implicit assumption here.

This comment is just a rehash of the increasingly outdated and incorrect assertion that LLMs can't possibly exhibit any creativity -- and it's also incorrect.

If you're yearning for "old skool artisanal weirdness of yore", look up the trend on Twitter a month or two ago of people asking Claude to make YTPs. They ended up very weird and artisanal in a way distinct from how any human would do it.

Okay, this is a pet peeve of mine, so forgive me if I come off a little curt here, but-- I disagree strongly with how this was phrased.

"Generative AI" isn't just an adjective applied to a noun, it's a specific marketing term that's used as the collective category for language models and image/video model -- things which "generate" content.

What I assume you mean is "I think <term> is misleading, and would prefer to make a distinction".

But how you actually phrased it reads as "<term> doesn't mean <accepted definition of the term>, but rather <definition I made up which contains only the subset of the original definition I dislike>. What you mean is <term made up on the spot to distinguish the 'good' subset of the accepted definition>"

I see this all the time in politics, and it muddies the discussion so much because you can't have a coherent conversation. (And AI is very much a political topic these days.) It's the illusion of nuance -- which actually just serves as an excuse to avoid engaging with the nuance that actually exists in the real category. (Research AI is generative AI; they are not cleanly separable categories which you can define without artificial/external distinctions.)

This is so strange, because, at a low level, a branch isn't even a "thing" in git. There is no branch object type in git, it's literally just a pointer to a commit, functionally no different from a tag except for the commands that interact with it.

> If another model can find the vulnerability if you point it at the right place, it would also find the vulnerability if you scanned each place individually.

They didn't just point it at the right place, they pointed it at the right place and gave it hints. That's a huge difference, even for humans.

> That said, a well-designed scaffold naturally produces this kind of scoped context through its targeting and iterative prompting stages, which is exactly what both AISLE's and Anthropic's systems do.

Unless the context they added to get the small model to find it was generated fully by their own scaffold (which I assume it was not, since they'd have bragged about it if it was), either they're admitting theirs isn't well designed, or they're outright lying.

People aren't missing the point, they're saying the point is dishonest.

No, you wouldn't. The vulnerability has been in the codebase for 17 years. Orders of magnitude more than 20k in security professional salary-hours have been pointed at the FreeBSD codebase over the past decade and a half, so we already know a human is unlikely to have found it in any reasonable amount of time.

> Their job literally depends on them finding Mythos to be good, we can't trust a single word they say.

TFA article is literally from a company whose business is finding vulnerabilities with other people's AI. This article is the exact kind of incentive-driven bad study you're criticizing.

Hell, the subtitle is literally "Why the moat is the system, not the model". It's literally them going, "pssh, we can do that too, invest in us instead"

> I mean, I literally posted a quote from the CEO of one of the two major companies saying that SWEs are 6-12 months away from being replaced.

Even ignoring the other ways you're misrepresenting the, there's a huge difference between "might be" and "are going to be".

I'm sorry if English isn't your first language, but we're going to have to agree on basic grammar or else it's not going to be productive for me to continue responding to the flaws in your argument.

One of the five fundamental pillars of conservative thought, as phrased by wikipedia (which is itself merely paraphrasing Russel Kirk, a foundational of post-war American conservativatism), is:

> A conviction that society requires orders and classes that emphasize natural distinctions.

Racism and bigotry are not errant additions to conservatism, they're a logical extension of one of its foundational pillars. (Though that is not to say that the left is not without its racism and bigotry as well, it's just less of a natural fit)

> We’ve been suggesting that programmers are going to be replaced by simpler programming languages, gui programming tools, no code tools, low code tools, and now AI.

The difference is that it's actually working this time. Non-programmers are writing full apps. Sure, they're simple ones, often just CRUD and UI, but it actually is changing things in a way it never has before. You can't assert something is the same as everything previous when there's already evidence that it's different.

> No - you don’t get to make wild predictions and say “oh I didn’t actually mean that, look how succesful we are though”.

Except that's not what's happening here. I'm criticizing you for misrepresenting what claim was made in the first place. No where in your evidence have you shown anyone "walking the claim back". If anything, TFA is claiming evidence of an LLM doing "most" of what SWEs do "end to end" three months ahead of schedule.

If you want to present evidence Dario (or another CEO -- I'm sure Sama has made much more fantastic claims that you could falsify) made claims that didn't pan out, be my guess, but don't tell falsehoods about the evidence you are presenting.

(And no, I'm not counting breathless tech reporters -- everyone knows how much to trust them when they report a cure for cancer -- they'll say everything is a miracle cure. But the fact that hundreds of "miracle weight loss cures" that never panned out made the new in the past several centuries didn't make GLP1s fake just because they had the same type of hype.)

> “I think… I don’t know… we might be six to twelve months away from when the model is doing most, maybe all of what SWEs (software engineers) do end to end.”

I think it's disingenuous (as disingenuous as you're accusing these marketing teams of being) to paraphrase that as "being told on one hand that we are 6 months away from AI writing all Code". It's merely stating that it's a real possibility. (It's also disingenuous to use a post complaining about a behavioral regression bug as evidence that it's not progressing)

Dismissing it as impossible is silly, considering how close it already is to a junior dev. Keep in mind that 14 months prior to that statement was before we even had any public reasoning models. Things really are moving that fast, it's just, at the moment, unclear how fast.

If you're on Pro, Group Policy tends to be more stable than plain registry tweaks.

In particular, I advise everyone to turn off web search in the start menu, it makes it so much faster and more useful:

   Computer Configuration > Policies > Administrative Templates > Windows Components > Search > Don't search the web or display web results in Search

> A human can sit down to play a game with unknown rules and write a spec as he goes.

Some humans can. Many, if not most humans cannot. A significant enough fraction of humans have trouble putting together Ikea furniture that there are memes about its difficulty. You're vastly overestimating the capabilities of the average human. Working in tech puts you in probably the top ~1-5% of capability to intuit and understand rules, but it distorts your intuition of what a "reasonable" baseline for that is.

In retrospect, it seems obvious that we hit AGI by a reasonable "at least as intelligent as some humans" definition when o3 came out, and everything since then has been goalpost moving by people who have higher and higher bars for which percentile human they would be willing to employ (or consider intellectually capable). People should really just use the term "ASI" when their definition of AGI excludes the majority of humans.

Edit: Here's the guy who coined the term saying we're already there. Everything else is arguing over definitions.

https://x.com/mgubrud/status/2036262415634153624

> Well, Lars, I INVENTED THE TERM and I say we have achieved AGI. Current models perform at roughly high-human level in command of language and general knowledge, but work thousands of times faster than us. Still some major deficiencies remain but they're falling fast.

> Yet I regularly catch them making errors that a human never would

I have yet to see a "error" that modern frontier models make that I could not imagine a human making - average humans are way more error prone than the kind of person who posts here thinks, because the social sorting effects of intelligence are so strong you almost never actually interact with people more than a half standard deviation away. (The one exception is errors in spatial reasoning with things humans are intimately familiar with - for example, clothing - because LLMs live in literary space, not physics space, and only know about these things secondhand)

> and which betray a fatal lack of any sort of mental model.

This has not been a remotely credible claim for at least the past six months, and it seemed obviously untrue for probably a year before then. They clearly do have a mental model of things, it's just not one that maps cleanly to the model of a human who lives in 3D space. In fact, their model of how humans interact is so good that you forget that you're talking to something that has to infer rather than intuit how the physical world works, and then attribute failures of that model to not having one.

Correct, but in practice this is almost never done, because the way the US legislative system is set up it's almost always more convenient to have judges rewrite laws instead of legislators.

In the US, the process of removing unjust laws generally involves violating them, so that courts have the opportunity to legislate from the bench.

Because silent mode is for the notifications. App volume has its own dedicated buttons.

HN already has something like that -- high-karma accounts can flag comments/posts which are a poor fit for HN. It's just a blacklist, not a whitelist.

Not every post is from the website of the person who is the topic of it. It's common to have e.g. a blogpost about $thing and then a new account chimes in with "Hey, I authored $thing 10 years ago when I was working for $company, someone linked me this post. [some contributions to the topic]"

I mean, to be fair, both things can be technically true. There can be lots of interesting things being done, even while most can be low-effort garbage.

But this is just Sturgeon's Law (ninety percent of everything is crap), not an actually insightful addition to the discussion, and I very much agree it's a stale take.

That's an extremely niche use case and not even a remotely selling point for probably 99% of buyers, much less the "biggest" one.

> This is genuinely

Well, I have a good guess which of those models is your favorite.

I'm not even saying that Claude wrote this - because it still reads as human written, and it's not badly written - but it has just enough Claude voice in it that it feels like the thing where humans inevitably start talking like the people (or simulacrums thereof) that they interact with most. (Heck, you did "It's not X it's Y" twice)

...Or maybe I'm the crazy one here. I don't know. But if I'm right, it's fascinating to see this happen.

Also by the people that just work there(, man).

I mean, as dumb as it is, there is a certain musicality to hearing someone with a southern accent sardonically call it the dee-oh-dubya.

The full[1] quote is:

> “Laws are a threat made by the dominant socioeconomic ethnic group in a given nation. It’s just the promise of violence that’s enacted, and the police are basically an occupying army, you know what I mean?”

...Which is funny, but technically speaking, it's (more or less) a paraphrasing/extrapolation of the very serious political science definition of a state, “a monopoly over the legitimate use of violence in a defined territory”

[1] Minus the last line, which I will allow others to discover for themselves

It's a little weird, too, because Claude definitely isn't the only one approved for use on classified systems in general; both Grok and OpenAI have models approved, at the very least.

https://devblogs.microsoft.com/azuregov/azure-openai-authori...

https://x.ai/news/government

They announced it almost a year ago:

https://www.anthropic.com/news/claude-gov-models-for-u-s-nat...

https://support.claude.com/en/articles/13756069-public-secto...