Episode 5 · 2026-07-31
AI Detection and the Presumption of Guilt
An AI detector's label damages a writer whether or not it is accurate, especially when the scan happens unannounced.
Guest Limited Edition Jonathan on Substack's AI detector: a label does its damage whether or not it is accurate, and a writer scanned without being told pays for it.
Host Sam Rogers · guest Limited Edition Jonathan, the mad-science builder and creative · 24 min
A choice this episode can help with
Does the label on this writing measure the thing your decision needs?
Run it on your own work: The Label Decision Card, one page, any model.
Watch it below, or read the conversation.
Listen
Or follow the show on Apple Podcasts, Spotify, or in any podcast app.
The signals
- Jonathan: the label does its work whether or not it is true. Substack shipped a button that reports whether a post is AI generated, AI assisted, or human written, and arguing about its accuracy concedes the wrong ground. His anchor is the nocebo case: New York announced fluoridation, the rollout ran late, nobody told the public, and on the announced day people called in with headaches and stomach aches. Nothing was in the water. A detector does not need to be wrong to do damage; it only needs to be believed.
- Sam: origin and quality are different properties, and only one of them is being measured. Textual origin, production process, and communicative quality are three separate things. Pangram estimates the first from a finished artifact. Disclosure lets an author testify about the second. Nothing on the platform touches the third, while the cost of producing plausible language collapsed and the methods for evaluating meaning did not move. Knowing an apple came from a local farm does not tell you whether it is a good apple.
The subtractions
- Jonathan: subtract the AI-assisted label, everywhere, including the voluntary kind. Not only Substack's machine-applied one. He held this position before the detector existed and holds it against self-disclosure too: journalism has never owed its audience its sources, its editors, or its ghostwriters, and volunteering the information is what teaches an audience there was something to be suspicious of. His line on the scan button: you are not entitled to my process. His last word on tape: engage with my ideas, not my process.
- Sam: retire authorship as a quality claim, in both directions. Human-written is not a quality claim and machine-flagged is not one either. Both are cheap proxies, reached for because they are easy. What is newly possible is naming what good actually means and measuring that directly, which neither the label nor the detector attempts. This is a genuine disagreement: Jonathan removes the signal, Sam keeps the signal and strips the quality meaning out of it. Neither talked the other out of it on air.
About this episode
Two voices, no cohost, and two recordings. The conversation was livestreamed on Wednesday while Limited Edition Jonathan was in a car on a two-hour drive to the emergency room where his son had been taken from summer camp. He knew that when the session started and kept the commitment anyway. The cell connection did not survive the trip. The episode is that recording, with host pickups filling the places where the connection took an answer mid-sentence, and the drops are left audible rather than smoothed. It is his argument with commentary, not a two-hander, and it is the first episode of the show assembled this way.
The clearest casualty is the question at 7:39: how does this actually hurt Substack and Substack readers? He gets one sentence in before the line goes. The answer, from his published piece: the flagged writer stops publishing. Not a debate, not a correction, just a vacuum. Stigma does not distribute by merit, it lands on whoever is least equipped to fight it, and his whole case for AI is that it lets a writer without a team compete against people who have one. Those are the writers a label discourages first.
The asymmetry is the part neither the tape nor most of the coverage reaches. The scan runs downward only: writers are scanned, the platform's own posts are not, and Chris Best's announcement of the feature carries no disclosure of its own. Writers are not notified, and the appeal arrives after the label already exists. A presumption of guilt with a customer service form attached.
Worth stating what Best actually argued, since the episode spends its length disagreeing with him. His piece defines the problem as deception, text passed off as human when it is not, which he and Freddie deBoer both call a con. On that definition the target is the faker. What shipped scans everybody and flags assistance of any kind at a hundred words or more, so it reverses hardest on honest, disclosed, assisted writing, which is the category his definition was never about. Best writes that when he wants Claude's opinion he will ask Claude; the button is a machine's opinion about whether a writer is worth reading.
The show streams to Substack and this episode's Substack surface is part of how it ships, so the argument is being made in the room it is about. Named once on air, not litigated.
This week's links:
- Nothing Was In The Water: Limited Edition Jonathan's piece, the premise of the episode
- Against Claudefishing: Chris Best's case for the feature
- Writing Quality After AI: Sam's article on origin, process, and quality as separate properties
- Limited Edition Jonathan on Substack
No sponsor this week.
Want to bring your own signal and subtraction? Find yours.
Transcript
Read the conversation
Limited Edition Jonathan: Substack has a button that you can click on now that tells you whether or not, the post is generated by AI or assisted by AI or written by AI, which is just a stigma-generating machine. And, uh, that's gross. I want people engaging with my ideas. I don't want people engaging with my process
Sam Rogers: Welcome to Signals and Subtractions. I'm your host, Sam Rogers. Last week, Substack shipped an AI detector, which means a writer can now be scanned by a machine, labeled by that machine and how it recognizes its own, and the writer is invited to file an appeal, but only after the fact. Today's guest is my friend Jonathan Edwards, AKA Limited Edition Jonathan. He's a filmmaker who teaches AI fluency, often by documenting his own failures in public, and who built AI cred to find out whether anyone's actually AI fluent. He's one of the top 10 rising voices in technology on Substack, and this week he wrote a piece called Nothing Was in the Water his position is whether the AI label is accurate is beside the point. The label itself is doing harm. We live streamed this on Wednesday via Substack while Jonathan was unfortunately in the car. Not the kind of recording quality we usually have here at Signals and Subtractions, but in this case it couldn't be helped. I'll let him tell why. It's legit. Anyway, for this episode, I'm going to step in on the places where the cell connection ate his argument. He was a champ to show up at all
Limited Edition Jonathan: it's nowhere cool. It, it is the hospital. no, my son's at his summer camp right now, and, uh, he got violently ill, so they are taking him to the emergency room, and it's a two-hour drive for me.
Put it to work
What decision are you making from an AI label, and what evidence would better support it?
The newsletter
Issue 62 · Sunday 2 August 2026 · sent on Substack and LinkedIn.
One signal, one subtraction, one analogy, in five minutes. It draws on this episode but stands on its own. Every issue.
The method from this episode is one page with its copy-paste prompt and schema: The Label Decision Card, linked under the decision above. Every takeaway.
Keep the next choice in view
One signal, one subtraction and one analogy every Sunday. Five minutes, no video required.
Optional. Free on Substack, no account needed. Already on LinkedIn? Subscribe there.