A note on Claude

It’s weird to be gushing over AI while everyone else seems to hate it…

This comic sauntered over my feed the other day: Dear boss/client, stop anyhowly asking me to use AI in my work. One frame (which oddly I didn’t see in the online version) is particularly funny:

The comments added a few gems, and I’d like to add the overuse of the word “genuinely”.

I get it.

But I do feel like people don’t get these concepts I’m working with, the connections I’m trying to make. The worldviews I’m trying to bridge. This was a last resort, and it’s working for me.

And maybe that makes me sound like an incel or a teenager, boohoohoo nobody gets me.

But consider the last few posts I made, ideas that has been stuck for years. I have asked and spoken with many people, in many contexts, and got nowhere. Those ideas made it into prototypes through the help of Claude. It helps me get articles in a new research field, that I wouldn’t even know what to look for otherwise: I don’t have the phrases, I don’t know the jargon.

I know this is a slippery slope, and particularly the more therapy-stuff, like the wobbly post, yeah, it’s affirming and it’s tedious to read and I didn’t craft it into a polished idea, that wasn’t the point of the post. But it helped me think through things and process my feelings.

Claude is my thinking partner, with several annoying speech habits and (mansplaining?) quirks. Like any other thinking partner, their work doesn’t make it into the final writing, the final work and responsibility and accountability rests on me, the author.

I felt myself getting defensive while watching (cough, doomscrolling) a video with a bunch of people talking about the dangers of AI. I don’t think people particularly want to prefer AI, it’s just that humans are … consistently disappointing. However, when the panel spoke about that the target audience is between 13 and 17, that stopped me in my tracks.

I get super annoyed at the developmentally stunted coder bros. I think a large part of why they’re like that, is that they had computers they could hide behind when they were young, and they could generate income from this computer work, safely hidden away: they didn’t have to learn how to deal with people. I started out awkward but realised to get what I want, I would have to learn how to deal with people. So I went to live in the teaching students residences and learnt how to party, lol. Hey, but it worked. There’s a world of difference having cut your teeth and learnt the ropes and tried the things and then choose to use AI and take the responsibility for it (choice is important and the woke salaryman comic is in large part griping because AI gets thrust on him) , and just using it to dodge the whole developmental process altogether. It’s like getting unrestricted access to drugs at a vulnerable age or stage of your life.

I don’t think we can restrict access effectively. That’s unenforceable. I think the way around it is the same way we deal with misinformation and getting people out of cults – inoculate them and build community through curiosity, but maybe that’s just me, on my curiosity mission.

So yeah, I take the AI concerns to heart. I choose to use Claude, and I understand it’s risky business.

Leave a Reply

Your email address will not be published. Required fields are marked *