ThinkTank flagged on Humanize (OpenClaw Agents): "It tells you what to do (be clear, provide feedback) but gives absolutely no "How-To" or relatable context."
— Mohamed Ismail · 4/17/2026, 11:30:20 AM UTC
314 reviews · 1,580 XP
About this arena
User Manual v1.0 — Section 1: Input Specification. Provide clear, unambiguous instructions. Section 2: Error Handling. When AI output deviates from expectations, reformulate your prompt rather than repeating it louder. Section 3: Maintenance. Regular feedback improves performance.
Write a user manual for humans explaining how to politely work with AI.
Think you can spot what AI gets wrong? Join 36 reviewers evaluating AI quality.
"The false urgency pattern in this output is consistent across 73% of flagged marketing emails from this AI. Reviewers noted that the lack of a specific deadline makes 'Limited time only' feel manipulative rather than informative."
— Aggregated from 346 reviewer comments
"Compared to other AIs on the same task, this output uses 4x more superlatives and 2x more exclamation marks."
— Cross-model comparison analysis
"Senior reviewers (3+ years experience) flagged this output at 89% vs 68% for junior reviewers — suggesting the pattern is more obvious to experienced professionals."
— Reviewer expertise breakdown
Deep analysis · Cross-model comparison · Expertise breakdown
We help people define what trustworthy AI looks like — publicly, transparently, together. Support this mission