Your intellectual fly is open when you use an LLM to author a post (2025) • Hacker News Board

Re: Your intellectual fly is open when you use an LLM to author a post (2025)

#251

Post by relevant_stats » Sun, Sep 06, 2026, 6:54 PM UTC

rahuldracula wrote: post #120

I don't agree. Good writing and good ideas are entirely different things. Just read all the beautifully written, inaccurate stuff by the NYT or the Guardian. Also, non-native English speakers have to use LLMs to share their views so that they are not judged on their writing. This post is also a good example; it's well written, and I enjoyed it. The idea could have been expressed simply as: "People can smell LLMs in y…

> I don't agree. Good writing and good ideas are entirely different things. Just read all the beautifully written, inaccurate stuff by the NYT or the Guardian.

I'd be very curious of some examples when great ideas are being presented in bad writing (or bad speaking for that matter).

> Also, non-native English speakers have to use LLMs to share their views so that they are not judged on their writing.

weird statement. If they use LLMs they will be then judged for both their inability to write in English and for their usage of AI. Not good.

Re: Your intellectual fly is open when you use an LLM to author a post (2025)

#252

Post by daveguy » Sun, Sep 06, 2026, 6:57 PM UTC

cweld510 wrote: post #202

bcherny wrote: post #166

Earlier quoted context omitted.

(I work at Anthropic) I agree. I use an LLM to write my code, but I do all of my writing by hand, since it helps me think. It reminds me of the transition over the last year from AI-assisted coding to AI doing all the coding. At first the code output wasn't good enough, and humans read and iterated on the code all day, so the details of the source code mattered. Now, the code is largely high quality and it meets a la…

Writing is already amenable to many different levels of abstraction, though. If an LLM can expand your outline into writing, then you aren’t writing at the correct level of abstraction in my opinion; you should instead be explaining how you arrived at your chosen outline. You don’t need to explain the details because any party can generate those with an LLM; same as how many PRs today can be auto-generated and no one…

This only applies if LLMs aren't making mistakes 20% of the time and that's the problem. When you're only saving time on the easy part, it doesn't matter if you're working twice as fast because review of the tricky parts is still going to take 80% of what it would have taken to do the whole thing. Total effort ends up being more rather than less if you want the same quality.

Re: Your intellectual fly is open when you use an LLM to author a post (2025)

#253

Post by tomrod » Sun, Sep 06, 2026, 7:07 PM UTC

bcherny wrote: post #166

jeremyjh wrote: post #62

These are all important points and I love the analogy. But there is an even bigger issue with having LLMs write for you: Writing is thinking. Thinking and deciding. There have been many times when I start out writing something substantial - could be an email, a blog post, a software design document, anything - when my own views substantially changed during the writing process. Writing forces you to serialize your tho…

(I work at Anthropic) I agree. I use an LLM to write my code, but I do all of my writing by hand, since it helps me think. It reminds me of the transition over the last year from AI-assisted coding to AI doing all the coding. At first the code output wasn't good enough, and humans read and iterated on the code all day, so the details of the source code mattered. Now, the code is largely high quality and it meets a la…

> wonder if long form writing will go the way of code.

I worry about AI Loopidity here though. Think about the similar analogy of email. If my set of ideas is condensable to bullet points, but I use AI to expand the content, then I add no information density and a lot of noise. Other folks then use AI to summarize the content to a list of bullet points, ideally the same but not certainly the same, and thus communication has been only partially successful.

Re: Your intellectual fly is open when you use an LLM to author a post (2025)

#254

Post by a2ff6eeb0 » Sun, Sep 06, 2026, 7:09 PM UTC

rogerrogerr wrote: post #238

a2ff6eeb0 wrote: post #226

Earlier quoted context omitted.

Sure, but I don't know what GCC's behavior is, and I don't vet behavior differences between compiler upgrades. As long as the output works, why does it matter that the black box is deterministic?

Because _someone_ has vetted the output of GCC. It’s used in flight-critical stuff. The closest thing we have to vetting LLMs is “whoa look, it escaped this sandbox, that’s prolly not great but it’s so cool!”

Sure, I manually test the output of the LLM. Manual testing is actually the main role for humans doing software engineering these days.

I wouldn't use it for flight control software yet, at least not without careful review, but most software isn't exactly critical. At the same time, I wouldn't trust flight control software that was only reviewed by humans, since AI is so much better at debugging.

We'll probably need humans in the loop for safety critical software for at least a year or two, before AI fully outpaces humans at generating correct code.

Re: Your intellectual fly is open when you use an LLM to author a post (2025)

#255

Post by jampekka » Sun, Sep 06, 2026, 7:10 PM UTC

jakelazaroff wrote: post #150

jampekka wrote: post #140

Earlier quoted context omitted.

I did think about this while writing, and I made the compromise to accept that someone will nitpick about it to hopefully drive the point better for those willing to read it charitably. QED?

What is the charitable interpretation? I am not being coy or sarcastic here; it is genuinely (forgive the claude-ism) unclear to me what your arguments are.

With the last line I tried to condense (and oversimplify) about these ideas:

Not all writing is thinking: That all, even a lot, of writing, or parts of writing, is such that it will develop one's thinking much. For example most stuff I have to write, the dozen emails a day, the funding application boilerplates, the reports are stuff that don't really need (or deserve) much thinking but they have to get written. And even in the writing that deserves attention, there is stuff like grammar and spelling and surface style that usually take quite a bit of time after the ideas have been written down already.

Not all thinking is writing: for many cases writing is not a particularly efficient way to develop one's thinking, and e.g. visualizations, math, coding, discussions etc can be a lot better.

Re: Your intellectual fly is open when you use an LLM to author a post (2025)

#256

Post by a2ff6eeb0 » Sun, Sep 06, 2026, 7:10 PM UTC

cozzyd wrote: post #241

a2ff6eeb0 wrote: post #194

Earlier quoted context omitted.

> but how about you and the team who will eventually read and do code review. Why are you reviewing AI code in detail? Do you also review the assembly output of GCC line by line?

For instructions you really care about, yes of course you review the assembly output! Usually when you're doing SIMD or want to check atomics are doing what you expect.

You can do that with LLMs for the parts you really care about too. The LLMs aren't regenerating the codebase from scratch every time, so the results stick around.

Re: Your intellectual fly is open when you use an LLM to author a post (2025)

#257

Post by arjie » Sun, Sep 06, 2026, 7:11 PM UTC

bcherny wrote: post #166

jeremyjh wrote: post #62

These are all important points and I love the analogy. But there is an even bigger issue with having LLMs write for you: Writing is thinking. Thinking and deciding. There have been many times when I start out writing something substantial - could be an email, a blog post, a software design document, anything - when my own views substantially changed during the writing process. Writing forces you to serialize your tho…

(I work at Anthropic) I agree. I use an LLM to write my code, but I do all of my writing by hand, since it helps me think. It reminds me of the transition over the last year from AI-assisted coding to AI doing all the coding. At first the code output wasn't good enough, and humans read and iterated on the code all day, so the details of the source code mattered. Now, the code is largely high quality and it meets a la…

One could imagine a universe where the agent fills in citations and supportive points and so on, or makes a more conclusive argument but it seems you’d get better results leaving that to read time if it’s a one-shot. Steering prompts etc. with tool use to bring in other sources etc of course change this entirely. And regardless, I doubt we will read content like that directly ever again. Agents will act as per-person highly specialized adapter layers for information transmission.

The entire point is what runtime you’re running your code on. A computer with any modern stack requires a lot of text for you to communicate “spin a square around on its center” to it. A human requires only that short string because they have a faster natural language interpreter.

Text meant for a human can communicate “spin a square around its center” much better than any code that mimics it. In some sense, all programming is boilerplate expansion because computers have (until now) been unable to be programmed with anything approaching natural language.

Re: Your intellectual fly is open when you use an LLM to author a post (2025)

#258

Post by jampekka » Sun, Sep 06, 2026, 7:15 PM UTC

alpinisme wrote: post #148

jampekka wrote: post #136

Earlier quoted context omitted.

Sure, but doing an interpretative dance or an abstract painting can make you reconceptualize the whole concept too. But we're not really pushing those tools as much as writing. Take for example a non-native writer of the language. I'm sure having to check up words from a dictionary may help to reconceptualize things, but I'm sure also that it's not often very efficient. And I think similar is going on for natives too…

It’s not clear that doing an abstract painting can make you reconceptualize in the way that reframing in other words does. The point about putting in other words is that you may stumble on a clearer, more tractable, more extensible framing. The kind of reframing an abstract painting does is very different, more like changing your attitude or way of looking. But I don’t think it ever leads to a place where you will su…

I'm not much of an artist, but I'd guess trying to paint a painting of how quicksort works, in a way someone else can grasp it too, can need a lot more reconceptualization than writing a description about it. Doing plots or diagrams, or even implementing an algorithm, for sure often need more thinking than writing a description.

Re: Your intellectual fly is open when you use an LLM to author a post (2025)

#259

Post by arjie » Sun, Sep 06, 2026, 7:20 PM UTC

p-e-w wrote: post #109

andy99 wrote: post #47

Earlier quoted context omitted.

This misunderstands what it means to be better at writing. Root cause here is that writing is how humans communicate ideas. An LLM written thing isn’t doing that, it’s something equivalent to copy/pasting a Wikipedia article. Even if the writing gets better, it will still be necessarily empty when expanding anything that isn’t completely specified. LLMs seem already to be pretty good at translating, where you already…

> An LLM written thing isn’t doing that, it’s something equivalent to copy/pasting a Wikipedia article. Nonsense. I’m not sure whether this was ever an appropriate description of what LLMs do, but either way, they have obviously moved way, way beyond that.

Absolutely agree. A modern agent is a sophisticated tool that can link multiple sources together to create a coherent piece of information hyper-specified for an audience. However, if humans have outlines that are expandable by LLMs it makes the most sense to do that as late as possible.

I don’t like reading AI writing either but I’m sure that’s a transitory period. Single prompt text expansions are unlikely to be useful because they’re late-bindable. You could give the original to me and I might be able to understand better.

But a series of steering prompts with various sources brought in is a different story. At that point it’s just a question of whether the agent can put together good information and their current inability to do so is unlikely to mean an inherent problem.

Some kind of UI affordance for this might help: with the agent emitting tags that allow for auto-folding or expansion in a way that allows both concise text and exposition when required by the reader. Mechanical sympathy, but for code executing on a human: good old human sympathy if you will

Re: Your intellectual fly is open when you use an LLM to author a post (2025)

#260

Post by qsera » Sun, Sep 06, 2026, 7:23 PM UTC

jeremyjh wrote: post #122

bbor wrote: post #100

Earlier quoted context omitted.

They are powerful tools but they do not have any human understanding - that isn't their optimization target. Ofc the rest is all right on, but I'd quibble with this specific idea. LLMs are absolutely targeted at modeling human understanding, which is the same faculty that contains what we call perception (!= sensibility) and intuition (!= rationality). It would be nice to train them to be completely alien from the gr…

We don't have a reward function for "human understanding". We reward the appearance of understanding. We define goals that we cannot conceive of reaching without something like understanding happening. There is something happening, but it is alien and counter-intuitive - it makes bizarre mistakes that betray it - and we don't know what it is. I'm pretty sure it is not human understanding.

> we cannot conceive..

That is it. We cannot concieve it because we are new to it. Just like we would think of Stackoverflow as intelligent if we are fresh off the jungle and are not aware of how Internet works. Because without know that, we cannot conceive how Stackoverflow can produce answers without it "understanding"