Twitch tested an ML-powered prompt system that detects potentially harmful chat messages before they are sent and asks users "Are you sure you want to send this?" — nudging them to reconsider without blocking the message outright.

Announced as an experiment rolled out in phases. According to Twitch's 2024 Annual Transparency Report for the New Zealand Code of Practice, the nudges resulted in a 9.26% reduction in toxic chat messages without significantly affecting user engagement, with 98.23% of nudged users returning to chat within 48 hours. As of the report date, the experiment was active for approximately 10% of users. Twitch's official statement: "We're rolling out an experimental prompt powered by machine learning that encourages users to pause before sending chat messages that are potentially harmful."