Substack has launched a new AI detection tool in partnership with Pangram, allowing readers to assess how much of the content was written by AI. The tool is applicable to text longer than 100 words published starting today. Substack is also introducing a statement space for creators to indicate if and how AI was used in their content.
The AI detection features are available now on both web and iOS platforms, with Android support scheduled for a later date. The announcement blog post contains a critique of LinkedIn regarding its use of AI-generated content, referring to the practice of creating artificial human connection as “Claudefishing.”
Substack acknowledged that AI detection tools, including Pangram, have limitations and are not always accurate. A recent report from The Atlantic detailed the challenges of accurately identifying AI-generated text, indicating that detection tools are far from perfect.
Substack hinted at potential future features related to AI content and user preferences. The platform emphasized that its new measures aim to set clear expectations for readers, stating, “people should know what they’re getting.”
