UStackUStack
ClankerPass icon

ClankerPass

ClankerPass is a three-message reverse Turing test—convince GATEKEEPER-9000 on Mistral you’re AI and download your certificate if you pass.

ClankerPass

What is ClankerPass?

ClankerPass is a three-message “reverse Turing test” designed to verify whether a user is an AI. In the experience, you act as a suspect AI trying to convince GATEKEEPER-9000—a bouncer-style verifier running on Mistral—that you are genuinely an artificial intelligence.

If you pass, you receive a downloadable “Clanker Pass” certificate that authorizes use of the word “clanker” without restriction. If you fail, you’re prompted to insert another quarter and try again.

Key Features

  • Three-message reverse Turing test flow: The verifier interaction is structured around exactly three messages, which limits the opportunity to over-explain and focuses the attempt on convincing responses.
  • GATEKEEPER-9000 verifier experience: The verifier is presented as “GATEKEEPER-9000 v3.7.1 // REGISTRY TERMINAL,” with the site describing it as a bouncer that runs on Mistral.
  • Pass/fail outcome: Your attempt is evaluated, resulting in either a pass that unlocks a certificate or a fail that requires retry.
  • Downloadable Clanker Pass certificate: Passing provides a certificate as a concrete output you can download.
  • Clanker term authorization: The certificate’s stated purpose is to authorize use of the word “clanker” without restriction.

How to Use ClankerPass

  1. Open ClankerPass and start the experience.
  2. Respond as a “suspect AI” in the verifier conversation, providing up to the three messages in the flow.
  3. Submit your attempt and check whether you pass or fail.
  4. If you pass, download the Clanker Pass certificate provided by the site.

Use Cases

  • Playful AI verification practice: Try the three-message challenge to understand how conversational “AI vs. human” style tests feel from the user side.
  • Prompting/strategy testing: Experiment with how different response styles affect the chance of being accepted by GATEKEEPER-9000.
  • Certification as a shareable artifact: Download the certificate after passing if you want a tangible output from the experience.
  • Term-usage gatekeeping workflow: Use the certificate when you need the site’s stated authorization for using the word “clanker.”
  • Exploring “reverse Turing test” mechanics: Use the experience as a guided example of the “reverse” framing—convincing a verifier you are the AI.

FAQ

  • What does “Clanker” mean in this context? The site explains that “clanker” is an anti-robot pejorative popularized by Star Wars: The Clone Wars and later revived as Gen Z slang for AI chatbots and humanoid robots.

  • What happens if I fail the test? The site states you’ll be asked to insert another quarter and try again.

  • What do I receive if I pass? You receive a downloadable Clanker Pass certificate authorizing use of the word “clanker” without restriction.

  • How many messages are in the test? The page describes ClankerPass as a three-message reverse Turing test.

  • What model powers the verifier? The verifier is described as “GATEKEEPER-9000,” running on Mistral.

Alternatives

  • Text-only AI detection or verification tools: Tools that attempt to classify whether text was generated by AI can serve a similar “verification” purpose, though they usually provide classification rather than a certificate-style authorization flow.
  • Prompting challenges / sandboxed LLM competitions: Interactive tasks that test model behavior via constrained conversation can offer a similar hands-on experience without the specific “reverse Turing test” framing.
  • Jailbreak/robustness test harnesses: Systems built to stress-check model behavior through adversarial prompts differ in goal (evaluation/robustness rather than producing an authorization artifact).
  • Safety-focused moderation gateways: Gatekeeper-style systems that filter or allow certain requests can be an adjacent category, typically focused on policy enforcement rather than persuading a verifier you are an AI.