Substack Launches AI Detection Tool

Substack is stepping into the growing conversation around artificial intelligence with a new feature designed to help its users distinguish between human-written and AI-generated text. The platform announced this week that it's rolling out an AI detection tool, a move that could significantly impact how content is perceived and trusted on the site.

This new capability is powered by Pangram, a company specializing in AI detection technology. The tool will be capable of scanning various forms of content on Substack, including posts, notes, replies, and comments, providing an estimate of how much of the text might have been generated by AI or created with AI assistance. This initiative aims to bring more transparency to the platform's content ecosystem.

For readers interested in utilizing this feature, the process is straightforward. Content longer than 100 words can be analyzed by simply selecting the 'Scan for AI text' option from the three-dot menu located in the top-right corner of a post. Currently, the tool is available across the web and the iOS app, with an Android launch expected soon. This phased rollout suggests Substack is carefully integrating the technology to ensure a smooth user experience.

The introduction of such a tool by Substack is a notable development in the ongoing debate about AI's role in content creation. As AI-generated text becomes more sophisticated and prevalent, platforms are facing increasing pressure to provide mechanisms for identifying it. Substack's decision to partner with Pangram signals a commitment to maintaining authenticity and providing readers with the information they need to critically evaluate the content they consume. It will be interesting to see how this feature evolves and how it is adopted by both writers and readers on the platform.