The AI is good vs. AI is bad thing is trite. The your AI process is good vs. bad thing is also - let's just say, lacking in an appropriate amount of nuance. And if you really want to get into it, which is worth it, you should get a bunch of users of various powers together for a conversation like we did over on Epsilon Theory Office Hours last week.

There's a lightly edited transcript up on Panoptica if you want a good sketch of what came out, but I wanted to drill into one particular aspect of it here.

First, you should know Adam Butler was the inciting incident for this call. After the whole Druckenmiller-used-AI-to-write-his-WSJ-op-ed freakout, fueled further by Druck saying yeah so what, you're dumb if you didn't, Adam emailed a group of us ET Unplugged contributors and said, seeing as his writing process currently does lean heavily on AI, he didn't need any similar chance of pushback in his life so he's taking a step back from writing in public for a minute.

Now, Adam's style of writing in general is not super prose-heavy or overtly emotionally expressive, and I've spent enough time talking with Adam to feel like I know his voice, and I've never once read Adam's stuff and thought well that's AI, not Adam.

And yet here's Adam saying AI is omnipresent in his whole process, he doesn't feel bad about it, and in actuality he feels quite wonderful about how he's using it - because it helps him organize and get everything assembled in a way no other tool can. But why continue to share his work in public if people are just going to auto dump on it?

I know the average AI-written sales pitch sucks. I also know anything Adam writes sounds like Adam even if he used AI to write it. Why does the average user's crappy copy and Adam's assisted copy feel so vastly different to me?

Brent Donnelly, who I think uses AI in a similar way to the way I use it, said this:

"My benchmark is always just: is the thing good? Yes or no. If I'm writing something that's from my heart, even if it's an essay, and there are no tricks in it - I wouldn't use AI because there would be no need to."

And then he described his actual process - taking the 200 words out of a thousand that don't feel right, asking AI for ten ways to say it better, and then either using one verbatim or going oh yeah, if I move the verb to the start of the sentence it sounds way punchier and rewriting it himself. The tool in service of the feeling he already knows is missing. He arrives at the same not-bothered conclusion I do.

Ben Hunt, who is staunchly opposed to using AI for anything creative, had this to say:

"The reason I spend so much time on my writing is that 90% of what I write, I throw away. I don't use it. Because in the writing, I'll say - no, that's not quite right, or that doesn't make sense, or my thinking was wrong on this. That is not what happens when you write with AI, because you tell AI 'here's my great idea,' and AI writes it, and you nod your head and go 'yeah, it's good.' You lose all of that editing process."

And he's right. About his process. The editing that AI won't do (I might have quibbles with this, but I know exactly the logic he's applying here, too) - the cutting, the throwing away, the discovering mid-sentence that you were wrong - that's where the bar Ben's set for himself lives. He's not saying Adam is wrong. He's saying his own process would be violated. You still have the same principles, but if you carry them to a different person and their process, you can get a different result.

It's the covenant - between how you think and how you feel - that I was describing in More Eccentrics, Less Influencers. Now I see it clearly, against all of these personalities.

Ben's covenant with himself doesn't support using the tool the way Adam's covenant supports him using it. No religious conflict is necessary because they both have a noble path they're carving out. I can stand outside and trust both of them on what they're doing. Until they give me reason to do otherwise, which I don't anticipate, but if they did, I can update this current feeling I'm feeling.

So how is it possible we are all at least mostly okay with the outcome of Adam's process but worried about how we feel about it applied to our own? The covenant isn't with the tool. It's not even with the reader. The covenant is with yourself, and the reader can be trusted to feel the difference.