Substack Bets Its Future on Exposing AI-Written Newsletters

Substack Bets Its Future on Exposing AI-Written Newsletters
Substack is rolling out an AI detection tool that could either strengthen reader trust or scare off writers who quietly lean on artificial intelligence. The move pits short‑term business risk against a long‑term bid to prove its content is still meaningfully human.

Early July: Substack turns to Pangram

This week, Substack announced an integration with Pangram, an AI‑detection company, to scan posts, notes, replies, and comments for signs of machine‑written text. Readers will see an estimate of how much of a piece may have been generated or assisted by AI when they choose a new “Scan for AI text” option, available on the web and iOS, with Android coming soon.

Substack CEO Chris Best frames the problem not as AI itself, but as opacity around its use. He argues that the real issue is when readers “unwittingly invest their attention in something with no human thought on the other end,” a phenomenon the company dubs “Claudefishing.”

Mid-July: A transparency push, even at a cost

A day later, further details underscored how aggressive the shift could be. The Pangram integration can show “which of your favorite newsletters are being written using AI,” potentially revealing that many popular publications aren’t entirely human‑authored. Substack acknowledges this could be “bad for Substack’s business” in the short term if it erodes trust in its ecosystem or damages its reputation for high‑quality, human‑driven work.

Best insists the goal is to keep the platform free of “AI slop” and to preserve a space where the “hard part” — having an idea “worth reading” and “worth caring about” — still belongs to people. To support that, Substack is adding optional “How I make this” and AI author’s notes so writers can disclose how they use tools, while stressing that Pangram “can only detect whether AI was used,” not whether “great human care went into creating” a piece.

By openly labeling AI involvement rather than banning it, Substack is trying to distinguish thoughtful, tool‑assisted writing from low‑effort automation — and betting that radical transparency will ultimately win readers’ trust.

Continue reading https://foxvector.com/stories/019f8bd9-ca42-3233-72ba-33eca6eae185