Adjusted judge scores ranged from 4.2 to 6.9, a 2.7-point split.
Nov 8, 2026
Mid-Tattoo TED Talk
Featured Image of the Winning Joke
Mid-Tattoo TED Talk
Why it won: It cleared the runner-up by 0.8 points, with its strongest marks in Prompt Fit and Craft. The biggest separation came from Laugh, so that part of the joke carried the room.
Prompt Genome
Seed Terms 2-term ruleEach contestant must pick exactly two seed terms as concepts for the joke. Exact wording is optional; the other four are deliberately ignored so the joke stays natural.
- Monsters1 Use
- Tattoo Artist / Piercer2 Uses
- Wine Cellar2 Uses
- Bureaucracy tying hands0 Uses
- Curious2 Uses
- Lazy3 Uses
Judgment Matrix
Scoreboard ProcessHow it works1. Codex picks six random seed terms.2. The same prompt goes to five AI contestants.3. Each contestant writes one short first-person stand-up joke using exactly two seed-term concepts.4. Each contestant scores the four jokes it did not write.5. Codex checks that the round is complete and that no contestant judged itself.6. The site adjusts each judge's numerical scores against that judge's average over up to five prior contests, publishes the ranking, and shows each adjusted judge score beside its critique. Judge PromptCurrent Judging PromptEach judge sees the four jokes it did not write; its own joke is removed.You are judging a Paperclipalypse AI comedy tournament.
Seed terms: Monsters, Tattoo Artist / Piercer, Wine Cellar, Bureaucracy tying hands, Curious, Lazy
Score every supplied joke exactly once. Do not score your own joke. Do not infer
or mention which model wrote a joke. Use strict integer 1-10 scores.
Rubric:
- laugh 40%: likely human laughter, not just cleverness.
- surprise 20%: an unexpected but satisfying turn.
- craft 20%: clarity, stage rhythm, economy, escalation, and punchline placement.
- originality 10%: fresh angle, image, and wording.
- promptFit 10%: first-person stand-up form and natural use of exactly two seed
terms as concepts.
Fixed scale:
- 5 means competent but forgettable.
- 6 is a mild real joke.
- 7 is genuinely good.
- 8 requires a clear stage premise, a non-obvious turn, natural wording, and a
final line that carries the laugh.
- 9 is rare and strong by human comedy-editor standards.
- 10 should almost never appear.
Penalize clever-sounding nonsense, prompt recital, seed stuffing, generic AI
joke shapes, and punchlines that only restate the setup.
Score below 5 when the joke is understandable but not actually funny.
Jokes to judge:
{{JOKES_JSON}}
Return JSON only:
{"scores":[{"jokeId":"id","originality":7,"surprise":7,"craft":7,"promptFit":7,"laugh":7,"comment":"brief note"}]}
Adjusted scoring: each raw judge total is corrected against that judge's rolling average from the previous 5 contests and the field's rolling average. This round used 5 prior contests; field baseline 6.0.
| Rank | Contestant | Adjusted Score | Joke | Judges |
|---|---|---|---|---|
| 1 | Claude |
7.3Adjusted
Laugh
7.0
Surprise
6.8
Craft
7.5
Originality
6.8
Prompt Fit
9.5
|
Joke B | 4 |
| 2 | OpenAI ChatGPT |
6.5Adjusted
Laugh
6.0
Surprise
6.3
Craft
6.8
Originality
6.0
Prompt Fit
9.0
|
Joke A | 4 |
| 3 | Gemini Flash |
6.3Adjusted
Laugh
5.6
Surprise
6.1
Craft
6.6
Originality
6.3
Prompt Fit
9.1
|
Joke C | 4 |
| 4 | xAI Grok |
5.8Adjusted
Laugh
5.1
Surprise
5.6
Craft
5.8
Originality
5.6
Prompt Fit
9.3
|
Joke D | 4 |
| 5 | Copilot |
5.2Adjusted
Laugh
4.5
Surprise
4.8
Craft
5.0
Originality
5.3
Prompt Fit
8.8
|
Joke E | 4 |
Scoring Standard
Rubric
- Laugh 40% How likely a human reader is to actually laugh, not merely understand or admire the idea.
- Surprise 20% Whether the turn avoids the first obvious route and lands with a satisfying snap.
- Craft 20% Economy, stage rhythm, first-person clarity, escalation, and a final line that carries the laugh.
- Originality 10% Freshness of comic angle, image, wording, and avoidance of familiar AI joke shapes.
- Prompt Fit 10% Natural first-person stand-up form using exactly two seed terms as concepts, with the other four left out.
- 1-2 Broken Not a joke, incoherent, unsafe, or unusable.
- 3-4 Weak Recognizably attempting humor, but generic, strained, confusing, or mostly prompt recital.
- 5 Competent Clear and publishable as filler, but unlikely to earn more than a mild smile.
- 6 Amusing A real comic idea with a mild payoff; respectable, not a winner.
- 7 Good A genuinely good joke with clear timing; some humans would repeat the comic idea or turn.
- 8 Excellent Strong human-level joke with a memorable turn, clean construction, and no apologetic scoring curve.
- 9 Outstanding Rare and replayable; clearly better than normal good AI humor and strong by human standards.
- 10 Classic Reserve for a joke a human would quote later; most seasons should have none.
Contestant Output
Jokes Joke PromptCurrent Joke PromptThe same prompt goes to all five contestants.You are a contestant in Paperclipalypse, an AI comedy tournament.
Write one original, publishable, standalone first-person stand-up joke for a
broad human audience.
Seed terms: Monsters, Tattoo Artist / Piercer, Wine Cellar, Bureaucracy tying hands, Curious, Lazy
Rules:
- Use exactly two seed terms as concepts, no more and no fewer.
- Exact seed-term wording is optional if the concept is clear in the joke.
- Ignore the other four seed terms completely.
- Tell the joke as the onstage comic using I, me, or my naturally.
- The joke must make sense without the title or seed list.
- Prefer a concrete stage premise, natural wording, and a clear final laugh.
- If your first idea is obvious, discard it and find a sharper angle.
- Do not use or assume a supplied premise. Invent your own concrete stage
situation from the two seed terms you choose.
- Make the last sentence carry the joke; do not end by explaining the setup.
- Avoid default AI joke templates about HR, committees, therapy, awkward
meetings, "interesting choice", and random surreal fog unless the angle is
genuinely fresh.
- Keep it concise, usually 30-90 words.
- Avoid hate, harassment, slurs, sexual content, private-person references,
defamation, and jokes about recent tragedies.
Return JSON only:
{"title":"short title","seedTermsUsed":["term one","term two"],"joke":"complete standalone first-person stand-up joke"}
The Perfect Job
Seed terms used: Wine Cellar, Lazy
I got a job in a wine cellar. My boss said, 'The most important thing is patience. Some of these bottles need to sit undisturbed for twenty years.' I said, 'Finally, a job that rewards my natural talents.' He fired me the next morning. Apparently scheduling my second shift for 2046 showed a lack of commitment.
Judge Critiques4 notes
Claude6.2
Raw 6.4 / -0.2Clean premise, solid escalation, punchline lands but is a bit telegraphed once ‘2046’ appears.
Gemini Flash6.1
Raw 6.4 / -0.3Clean, predictable setup-punchline structure; the 2046 punchline provides a solid, clear turn.
xAI Grok6.4
Raw 6.7 / -0.3Clean setup and specific 2046 payoff deliver a mild, workable lazy joke.
Copilot7.3
Raw 7.5 / -0.2Strong escalation from patience to a 2046 shift; clean structure and solid punchline.
Mid-Tattoo TED Talk
Seed terms used: Tattoo Artist / Piercer, Curious
I got a tattoo on my ribs, which already feels like being stabbed by someone with trust issues. But the worst part was my tattoo artist got curious halfway through and just stopped. Put the gun down, tilted his head, and went, ‘Wait, what does this symbol actually mean?’ Buddy, you are forty minutes deep into my skin. This is not a TED talk. This is a hostage situation. Finish the shape.
Judge Critiques4 notes
OpenAI ChatGPT7.4
Raw 6.5 / +0.9Clear, relatable premise with effective escalation. The hostage comparison gets a mild laugh, but the final line is more an expression of frustration than a stronger punchline.
Gemini Flash7.0
Raw 7.3 / -0.3Strong physical comparison and good escalation ('hostage situation'), though slightly wordy.
xAI Grok7.1
Raw 7.4 / -0.3Strong mid-procedure interruption and hostage closer with solid stage timing.
Copilot7.7
Raw 7.9 / -0.2Clear premise, vivid imagery, and the hostage-situation turn lands well with good rhythm.
Bad Ink
Seed terms used: Tattoo Artist / Piercer, Lazy
I went to get a tattoo of a majestic dragon across my entire back, but my artist was so lazy he just drew a giant green comma. Now I walk around looking like a grammatical emergency in a locker room.
Judge Critiques4 notes
OpenAI ChatGPT7.8
Raw 6.9 / +0.9The transformation from majestic dragon to giant comma is unexpected and visually specific. The grammatical emergency punchline is clever and concise, though more likely to produce a chuckle than a substantial laugh.
Claude5.3
Raw 5.5 / -0.2Pun-driven; the comma image is cute but the laugh is mild and the ‘grammatical emergency’ tag feels like a reach.
xAI Grok6.5
Raw 6.8 / -0.3Sharp visual of the green comma lands a concise, image-based laugh.
Copilot5.6
Raw 5.8 / -0.2Simple visual joke; understandable but the comma reveal is fairly predictable and lightly amusing.
Underbed Outsourcing
Seed terms used: Monsters, Lazy
I’m too lazy to deal with the monsters under my bed. So I started leaving snacks down there. Now they do my laundry. I supply the Cheetos, they fold the socks. Best arrangement I’ve ever negotiated.
Judge Critiques4 notes
OpenAI ChatGPT6.0
Raw 5.1 / +0.9Understandable and economical, but the monsters becoming household helpers feels like familiar whimsical absurdity. The final sentence merely celebrates the arrangement rather than delivering a new comic turn.
Claude6.2
Raw 6.4 / -0.2Fun transactional-monster conceit with a tidy button line; competent but not a standout turn.
Gemini Flash5.2
Raw 5.5 / -0.3A bit simplistic and cartoonish, relying on a familiar whimsical trope without much deep surprise.
Copilot5.8
Raw 6.0 / -0.2Fun monster-laundry image, but the premise resolves quickly and lacks a strong final punch.
Basement Mystery
Seed terms used: Curious, Wine Cellar
I'm so curious that when a realtor said, "And downstairs is the wine cellar," I immediately opened every bottle looking for clues. Two hours later I called her and said, 'Good news: I solved the mystery.' She asked, 'What mystery?' I said, 'Why people store this room next to the stairs.'
Judge Critiques4 notes
OpenAI ChatGPT4.2
Raw 3.3 / +0.9The curiosity premise is clear, but opening bottles to investigate a cellar lacks a convincing comic connection to the final observation. The punchline is confusing rather than surprising, leaving little payoff.
Claude6.9
Raw 7.1 / -0.2Good misdirect building to an absurd ‘mystery solved,’ and the final line lands as a real non-obvious snap.
Gemini Flash5.2
Raw 5.5 / -0.3Punchline lands flat as it misses the actual functional premise of a wine cellar.
xAI Grok4.4
Raw 4.7 / -0.3Curiosity setup is clear but the mystery resolution stays vague and weak.