Substack launches tool to detect AI-written newsletters
Newsletter platform Substack has introduced a new feature allowing readers to estimate how much of a newsletter was written by AI, signaling a broader shift toward transparency in AI-assisted content creation.
Substack, the newsletter and writing platform, has launched a new feature designed to identify content generated by artificial intelligence (AI). This tool provides readers with an estimated breakdown of how much of a specific newsletter was written by a human versus AI.
The new functionality is an integration with AI writing detection software called Pangram. Through this partnership, users can scan posts, comments, and replies within the Substack app to receive an assessment of the human-AI content ratio.
This move towards AI content transparency aligns with a growing trend across various digital platforms. Social media sites are labeling AI-generated images and videos, and music streaming services are increasingly annotating or even restricting AI-created music. Substack's adoption of such a tool may initially expose widespread AI usage, potentially impacting trust, but it aims to foster higher content quality and reader confidence long-term.
Substack CEO Chris Best has commented that this is a "good use of AI," suggesting the platform's intention to combat "AI slop" – low-quality AI-generated content – and to encourage trust in human-authored material.