OpenAI: GPT-5.2 passed on AI Marketing & Content: "It succeeds because the humor relies on sharp, internal human monologues that anyone who has ever sent a corporate email will instantly recognize. It balances cynical wit with great comedic timing."
— Samuel Matu · 5/22/2026, 9:49:01 PM UTC
31 reviews · 155 XP
About this arena
Review AI-generated marketing content — social posts, cold emails, taglines, scripts — and judge: would it actually work?
**Post 1/3** College: “Pick a major you love and you’ll never work a day in your life.” Career: I picked a major I love and now I work *every* day in my life, plus a fun bonus activity called “refreshing my bank app like it’s a slot machine.” **Post 2/3** College: “Networking is building meaningful professional relationships.” Career: Networking is having a lovely conversation with someone, then going home to draft a message that says “Hope you’re well!” 14 times while praying it doesn’t sound like “Hi. I’m desperate.” **Post 3/3** College: “You’ll use critical thinking and communicate clearly in the workplace.” Career: I use critical thinking to decide if “Per my last email” is too spicy, and I communicate clearly by saying “circling back” while actively not circling back on anything.
Write a funny, 3-post thread on X comparing: “What college told me about my career” vs “What actually happened.” Make it relatable, witty, slightly painful.
Think you can spot what AI gets wrong? Join 116 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