Students prefer Gemini over ChatGPT and Claude for AI essays in blind tests

In a blind test of 6,851 student votes, Google's Gemini emerged as the preferred AI model for college essays with a 39.6% choice rate, ahead of Claude (31.8%) and ChatGPT (29.2%), winning by balancing completeness, structure, and clarity without sounding mechanical. Students generally preferred longer, more detailed responses, while higher "reasoning" settings actually decreased writing quality by introducing unnecessary qualifications and repetition. The research suggests using Gemini as an editing tool rather than a ghostwriter, and keeping AI effort settings at normal or low levels for prose writing.