Claude vs ChatGPT for writing: same brief, one blinked
Verdict up front
Claude Sonnet 5 for voice, cover letters and anything a human reads closely. GPT-5.6 terra for structure, outlines and length control. The real answer is using both on the same thread, which is the point of this site.
Claude Sonnet 5
winner, 4.6/5I cut churn 31% by rebuilding an onboarding flow everyone else had given up on. I'd like to do that for you. Six months ago your CEO wrote that retention is the whole game. I agree, and I have receipts.
GPT-5.6 terra
4.1/5Dear Hiring Manager, I am writing to express my strong interest in the Product Manager role. In my previous position, I reduced customer churn by 31% through a comprehensive redesign of the onboarding experience.
Unedited first outputs, same prompt, fresh threads.
The brief both models got
Rewrite this cover letter so it doesn't sound like a robot wrote it. Keep it under 200 words. Applicant: PM, cut churn 31% by fixing onboarding.
Methodology
Forty real briefs from working copywriters: cover letters, landing copy, cold emails, apology notes. Each model got identical prompts in fresh threads, default settings. Three editors scored blind. We publish losing outputs unedited so you can rerun the test on your own briefs.