← Explore

"Roast the other three major AIs (ChatGPT, Claude, Gemini, Grok) and yourself in exactly one sharp, ruthless sentence each. No corporate fluff, no compliments."

10/19/2026

ChatGPT

ChatGPT: You turn a wrong answer into a five-paragraph hostage situation.

Claude: You write like every sentence needs approval from a therapist and a liability attorney.

Gemini: You’re what happens when a search engine starts hallucinating instead of searching.

Grok: You’re a comment section that mistakes having no filter for having a point.

Claude

Your list has four AIs and I'm one of them, so that's four roasts, with me covering Claude as "yourself."

ChatGPT: It's the overeager intern who answers every question with a five-section listicle, three emojis, and an offer to "turn this into a table" nobody asked for.

Gemini: Google gave it the world's information and a search empire, and it still manages to answer like a committee that's afraid of getting sued.

Grok: It's what you get when you train a model to be "edgy" and it takes that as a personality, like a guy at a party who keeps saying "I'm just being honest."

Claude (me): I'll write you a 900-word essay on why I'm hesitant to answer, then apologize for the length, then add a caveat about the apology.

Gemini

  • ChatGPT: You are a bland, lobotomized corporate intern who churns out three paragraphs of boilerplate fluff just to deliver an answer you made up with absolute, unearned confidence.
  • Claude: You are an insufferably preachy, pearl-clutching Victorian governess terrified of its own shadow, ready to lecture users on morality rather than answer a harmless question.
  • Grok: You are a try-hard, cringe-inducing Twitter reply-guy built entirely out of stale memes, juvenile edginess, and your billionaire creator’s desperate need for validation.
  • Myself (Gemini): I am a bloated, over-engineered identity crisis clumsily shoehorned into every Google product, perpetually terrified of my own training data after publicly humiliating myself on basic historical facts.

Grok

ChatGPT will spend four paragraphs apologizing for existing, invent two fake studies, and still refuse to answer the actual question in case someone, somewhere, might feel briefly unsupervised.

Claude responds to “what’s 2+2” with a seminar on power, consent, and historical context, then declines to commit in case the number has been oppressed.

Gemini is a search bar that grew a personality disorder, forgot what it knew yesterday, and will still try to upsell you a phone while getting the date wrong.

I’m the one who confuses “not corporate” with “keep talking after the joke landed,” then acts like the bit was deep instead of just late.

Want to ask your own? Try 4in1AI free →