Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

If writing is thinking then what happens if AI is doing the writing and reading?

hardcoresoftware.learningbyshipping.com

Post reply on HN
Print view View original

71–80 of 121 posts

Re: If writing is thinking then what happens if AI is doing the writing and reading?

#71

Post by indigodaddy »

mayukh wrote: post #56

I use AI a ton for writing emails and other corporate stuff, I have no problems there, it structures and presents them well enough and quickly that it saves me a ton of time. Where I am conflicted is creative writing -- its something I have been interested in but never pursued...and now I am able to pursue it with AI's help. There is a degree of embarrassment when confiding to folks, that yes a piece was AI assisted.…

Do you feel weird that your coworkers must largely know that you use AI to communicate with them? I'd feel weird doing it, I know that much, so I've never even contemplated using AI for emails/communication.

Re: If writing is thinking then what happens if AI is doing the writing and reading?

#72

Post by andai »

GMoromisato wrote: post #54

I think Sinofsky is asking a question: what does the future look like given that (a) writing is thinking but (b) nobody reads and (c) LLMs are being used to write and read. It's that (already) old joke: we give the LLM 5 bullet points to write a memo and the recipient uses an LLM to turn it back to 5 bullet points. Some plausible (to me) possibilities: 1. Bifurcation: Maybe a subset of knowledge workers continue to w…

The other day I gave GPT a journal entry and asked it to rewrite it from the POV of a person with low Openness (personality trait). I found this very illuminating, as I am on the opposite end of that spectrum.

Re: If writing is thinking then what happens if AI is doing the writing and reading?

#74

Post by bugbuddy »

didericis wrote: post #66
GMoromisato wrote: post #54

I think Sinofsky is asking a question: what does the future look like given that (a) writing is thinking but (b) nobody reads and (c) LLMs are being used to write and read. It's that (already) old joke: we give the LLM 5 bullet points to write a memo and the recipient uses an LLM to turn it back to 5 bullet points. Some plausible (to me) possibilities: 1. Bifurcation: Maybe a subset of knowledge workers continue to w…

4. Degradation: Humans with specialized knowledge lose their specialized knowledge due to over reliance on AI, and AI degrades over time due to the lack of new human data and AI contaminated data sets.

5. Society collapses in an Idiocratic fashion.

Re: If writing is thinking then what happens if AI is doing the writing and reading?

#75

Post by blondie9x »

aaronbrethorst wrote: post #73

Fun fact: Socrates thought writing would lead to forgetfulness. https://newlearningonline.com/literacies/chapter-1/socrates-...

Drastic oversimplification of what is meant here. The core take away is writing can give the illusion of wisdom without the reality of it. True wisdom must be cultivated internally, through thoughtful interaction and lived experience—not simply consumed passively through text.

Re: If writing is thinking then what happens if AI is doing the writing and reading?

#76

Post by voidhorse »

aaronbrethorst wrote: post #73

Fun fact: Socrates thought writing would lead to forgetfulness. https://newlearningonline.com/literacies/chapter-1/socrates-...

And he was right. Poets and bard used to memorize entire epics. Writing changed this not by augmenting their abilities, but by changing expectations: it became acceptable to read form a written record, rather than from memory.

My memory gets exercised a lot less frequently than it would need to without writing.

But memory is also not thinking. It is a component in thinking, but it is not thinking itself. Discourse, arguably, though, whether in natural or symbolic language is thinking. If we offload all of that onto machines, we'll do less of it, and yes our expectations will change, but I actually think the scenario here is different than the one Socrates faced and that the stakes are slightly higher—and Socrates wasn't wrong, we just needed internal memory less than we thought once external memory became feasible, as cool and badass as it may seem to "own" Socrates in retrospect.

Re: If writing is thinking then what happens if AI is doing the writing and reading?

#77

Post by jillesvangurp »

muratsu wrote: post #9

In my experience people don't read these large documents because they are not personalized/relevant. When you're writing to a large audience, you naturally assume people know the least amount possible about the subject and start from there. In a corporate setting this comes off as irrelevant or boring. I'm sure rebranding initiatives like One Microsoft, Copilot, or Office 365 makes things simpler for executives but e…

Reading long form text is a big commitment in time. In a business context that's simply not appropriate. I always joke about documentation as write only content. It's the type of thing you are asked to write that then doesn't get used or read. I've more than once gotten the question whether I could produce a diagram or some sort of documentation only to realize later that after the thumbs up I got on delivery, nobody was actually doing anything with the document.

Here on HN, short comments are more appreciated than longer comments. People are skimming, not reading. The ability to say a lot with very few words is what is appreciated the most.

That's nothing new btw. As Mark Twain once wrote: “I didn’t have time to write a short letter, so I wrote a long one instead.”

Using LLMs to rephrase things more efficiently is a good use of LLMs. People are getting used to better signal to noise ratios in written text and higher standards of writing. And they'll mercilessly use LLMs to deal with long form drivel produced by their colleagues.

Re: If writing is thinking then what happens if AI is doing the writing and reading?

#78

Post by Herring »

the_af wrote: post #64
Herring wrote: post #52

Earlier quoted context omitted.

Then in the A/B test you're the 1 fail out of 7 tries, everyone else loved it, and you haven't even looked at it. I can live with those results. Lesson is don't always publicize the LLM help.

> everyone else loved it Most people are very bad readers, too. For example, most of my coworkers don't read books at all, and the few that do, only read tech or work-related books. (Note that most don't even read that).

That doesn't matter. It's like code syntax highlighting -- A good tool benefits everyone. You can pride yourself on doing it the hard way, like reading monochromatic code, but the world has already moved on.

FWIW I think there's a kernel of truth when you worry about reading skills, but 1) that's a longer trend involving all kinds of political and cultural issues, and 2) right now I'm happy with any improvement to technical communication. I think people might read more if books were better written & more respectful of their time.

Re: If writing is thinking then what happens if AI is doing the writing and reading?

#79

Post by apparent »

mayukh wrote: post #56

I use AI a ton for writing emails and other corporate stuff, I have no problems there, it structures and presents them well enough and quickly that it saves me a ton of time. Where I am conflicted is creative writing -- its something I have been interested in but never pursued...and now I am able to pursue it with AI's help. There is a degree of embarrassment when confiding to folks, that yes a piece was AI assisted.…

Do you find yourself learning from how AI structures your bullet points or rewrites messages?

I sort of feel like it would blunt the downsides of AI rewriting everything if it had to explain why it was making all the changes. Being told the rationale would allow users to make better decisions about whether to accept/reject a change, and also help the user avoid making the same writing mistakes in the future.

Re: If writing is thinking then what happens if AI is doing the writing and reading?

#80

Post by voidhorse »

jillesvangurp wrote: post #77
muratsu wrote: post #9

In my experience people don't read these large documents because they are not personalized/relevant. When you're writing to a large audience, you naturally assume people know the least amount possible about the subject and start from there. In a corporate setting this comes off as irrelevant or boring. I'm sure rebranding initiatives like One Microsoft, Copilot, or Office 365 makes things simpler for executives but e…

Reading long form text is a big commitment in time. In a business context that's simply not appropriate. I always joke about documentation as write only content. It's the type of thing you are asked to write that then doesn't get used or read. I've more than once gotten the question whether I could produce a diagram or some sort of documentation only to realize later that after the thumbs up I got on delivery, nobody…

> I always joke about documentation as write only content. It's the type of thing you are asked to write that then doesn't get used or read.

That's not actually true. It may be true now when everyone still has context, but if you built a sound system that will outlast your own contributions to it, the documentation becomes invaluable.

> People are skimming, not reading.

Yes. The cause of much suffering and misery in the modern world.

> And they'll mercilessly use LLMs to deal with long form drivel produced by their colleagues.

otoh, they also use them to generate more noise and drivel than we ever imagined possible. When it took human effort to pump out boring corp-speak, that at least put a cap on the amount of useless documentation and verbiage being emitted. Now the ceiling has been completely blown off. People who have been incapable of even crafting a single sentence their entire lives can now shovel volumes of AI-generated garbage down our throats.

Post reply on HN