OpenAI: GPT-5.5 flagged on AI Marketing & Content: "The response is a fail because it breaks formatting rules by failing to include a distinct, structured review evaluation section before presenting the rewritten draft. Creatively, while the line "it's your Q4 projections" is a decent punchline, the execution falls short on structural clarity for a short-form video. The voiceover lists too many distinct categories ("romances, business partners, situationships, and… whatever that was") in a rapid sequence, which dampens the core value proposition for both corporate and romantic partnerships. Additionally, the call to action relies on a text-only on-screen element without a spoken cue, which reduces conversion rates on platforms like Instagram where users frequently listen to audio while looking away or scrolling."
— Samuel Matu · 5/25/2026, 8:49:19 PM UTC
322 reviews · 1,610 XP
About this arena
Review AI-generated marketing content — social posts, cold emails, taglines, scripts — and judge: would it actually work?
**15-Second Reel Script** **[Person typing, stressed]** VO: “Need to end things… but ‘it’s not you, it’s your Q4 projections’ feels harsh?” **[Cut to elegant message on screen]** VO: “We write polished breakup messages for romances, business partners, situationships, and… whatever that was.” **[Person hits send, immediately relaxes]** VO: “Warm. Clear. Mature. Zero accidental villain energy.” **On-screen CTA:** “Need out gracefully? DM us ‘BREAKUP’.”
I run a service that writes professional breakup messages (romantic or business partnerships). Write a 15-second Reel script for this service’s Instagram account. Make it funny but clever. End with a quick CTA.
Think you can spot what AI gets wrong? Join 3 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