yaelwrites

I'm a pretty late adopter and nost of my LLM use has been

1) vibe red teaming (having Codex RE the data flow of various extensions, comparing the attestation with the ToS, seeing if I think it's a potential FTC violation) which is just for fun and I do nothing with,

2) feeding tons of documentation into NotebookLM and asking if everything in a short document is attributable (often when it says it's not, it is, but you have to kind of read between the lines). I think this is a good use of AI if rigorously fact-checked.

3) feeding 20-30 job descriptions + cover letters, asking for outlines for new cover letters. For this one I think it may be faster to go back to creating templates because I have to entirely rewrite things anyway. (It does add lies! And conflates things! V bad for a cover letter!) But I think use of machines to write letters being read by machines is pretty justifiable.

4) I've been experimenting with Claude editing. I think what bugs me about people assuming everything is LLM-written is that it's rather dismissive and difficult to tell the difference between "a human spent X hours on this and made some stylistic tweets" vs "a machine wrote this in response to a prompt."

I do find myself immediately scrolling away too, though. A friend had a LI post for a course he's teaching and I could tell it was AI because it's not how he writes. That said, the post performed phenomenally well, which is probably more important for selling a course than my opinion is.

I'm also not confident in my ability to distinguish between 1) something an AI write from scratch, 2) something he may have iterated with an AI, feeding it many different versions of a post, going back and forth, etc. I think both would show up as AI in the extensions I've seen, but I also think they're very different.

My thought process on it is complicated. I think I'm more disappointed than anything. It made me sad because his human writing is so unique, and I'd hate to see it all outsourced. But I also don't have the same "you didn't even bother to write this so why should I read it" reaction so many do, given that I don't know how much time he spent on it or the reasoning.

Mostly I worry that if the whole of human knowledge becomes AI slop, and AI is then being fed that slop, that the writing would get even worse.

Re: fact-checking, I am far too paranoid/obsessive/whatever about accuracy to not review changes! Even NotebookLM is weirdly drawn to the first document you feed it, so if you have a second contradictory document, it seems to want to tell you why it's wrong. To be fair, I know people like this, too. The first version of a story they hear is what they will forever believe, even if every other version contradicts it.

Oh, no problem! But now I am thinking I may need to rethink using AI for light editing. I've gotten in the habit of that because I don't have a human editor for my blog, and I like getting that extra polish sometimes, but it's probably not worth it if it's going to make people not want to read.

Heh, no problem. I will say the rule of three is something I've been taught over and over again... in speaking training, in a class I took on fundraising, and most recently in a book I read. (See, I just did it again!) Book was by Mehdi Hasan.

I also recently reread a blog post I wrote a decade ago, and it had the "It's not X, it's Y" pattern over and over again.

There are a lot of parallels between "just ask an LLM" in response to a thoughtful, well-researched question and spurious accusations of heavy LLM use that never happened in response to a post, so that's actually a pretty astute point, just not the one I think you were trying to make.

And yet, here we are. :)

Good post. I'm trying to decide if this is better or worse than people sending their AI Q&A without anybody asking (as Cory Doctorow wrote about here https://doctorow.medium.com/https-pluralistic-net-2026-03-02...).

Wrong, those were all me. (I just checked the draft.) The only part I tweaked was the conclusion, plus I broke up a few paragraphs that were too long.

I often leave out details and specifics or make things purposely vague because I am writing about real people.

I wrote this by hand but I did run it through an AI for editing. The changes were pretty minor; just a couple of lines.

Yeah, that's why I described it as an examples of a bad design choice that may be accidental rather than a dark pattern. I do still think it's misleading, though.

See http://darkpatterns.org/british-airways-distract-from-cheape....

1. It's too bad that Switzerland has a mutual legal assistance treaty relationship with the United States which requires it to hand over any information legally available to their local authorities to the requesting government.

Here's a deep dive into ProtonMail's claims on Wired (disclosure: I wrote it) www.wired.com/2015/10/mr-robot-uses-protonmail-still-isnt-fully-secure/

I use Thunderbird with Enigmail, but for some reason it wants to load up thousands of archived emails and crashes my computer. It is annoying as fuck.

While I really enjoy reading and sending encrypted emails, and think it's far better than sending sensitive data in the clear, I am surprised to learn that there are people who use PGP and don't hate it or have issues with it.

I got accepted in NYC, had to turn it down because I'd already decided to go to a conference (ironically also in NYC) since I'd assumed I'd be rejected, emailed my cancellation and was told this wouldn't DQ me for SF, and then got rejected. We should have a meetup for all of the rejects, lol.

Glad to see this being discussed.

Relevant: http://www.inc.com/magazine/201309/jessica-bruder/psychologi...

I'm so psyched to see this. It's incredibly heartening to see YC set a great example because of their visibility and the amount of respect we all have for them in the tech/startup community. As someone who's unsuccessfully worked on getting codes implemented, I think this is absolutely huge and predict others will do the right thing and follow suit.

I agree that this is pretty vague. I don't think the more nitty gritty details (how to make an incident report, who's responsible for handling it, etc.) have to be posted publicly, but hopefully the actual policy is being worked on behind the scenes.

I just have to say how nice it is to read a discussion about this topic that's productive and supportive. I guess the new H/N algorithm that I whined about is actually working. :)

Great, once I get 991 more karma points I'll be welcome into the conversation.

I like this. It's all about positioning. I'm technically a freelance writer/editor, but I've found that when I call myself a "content strategist," I can command much higher wages doing the exact same work (since I have to do social, deep analytics, etc. as a writer anyway).

Science Daily (http://www.sciencedaily.com/) is a pretty good place to scan headlines, and they link to the original sources if you wan to get to the nitty gritty.

Thank you.

I actually work remotely, but thanks! He is great to work with. Super nice guy, too.

    −

This is cool. One addition: a great place to work out for those who like throwing heavy shit around is Catalyst Athletics in Sunnyvale. http://gym.cathletics.com/ (Disclosure: I work with the monthly journal they put out--but only b/c they're awesome.)