• Perle Labs Secures $17.5M in Funding
  • Season 1 is live!
  • Perle Labs Secures $17.5M in Funding
  • Season 1 is live!
  • Perle Labs Secures $17.5M in Funding
  • Join The Beta Launch
  • Perle Labs Secures $17.5M in Funding
  • Join The Beta Launch
  • Perle Labs Secures $17.5M in Funding
  • Join The Beta Launch
  • Perle Labs Secures $17.5M in Funding
  • Join The Beta Launch
Teach AI to spot unclear instructions before acting on them.

Workplace instructions often contain hidden ambiguity, missing timing, unclear responsibility, vague success criteria. For a human, these gaps might prompt a clarifying question. For an AI system acting autonomously, it can lead to confusion or unsafe action.

AI models need to learn when instructions aren't clear enough to execute reliably. That judgment requires human input.

The Ambiguous Instruction Identification Quest is now live on Perle Labs! Visit app.perle.xyz to start earning points.

What This Task Is

The Ambiguous Instruction Identification Quest is a set of 10 text identification tasks. Each prompt presents a short workplace-style passage containing multiple sentences — but only one sentence is vague or underspecified in a way that could cause confusion.

Your job is to identify the ambiguous sentence by highlighting it. These classifications train AI systems to detect unclear instructions before acting on them.

What You'll Do

Step 1: Read the passage carefully.

Step 2: Identify the sentence that contains an instruction or directive with missing information (unclear timing, conditions, responsibility, or success criteria).

Step 3: Highlight the sentence containing the ambiguity.

Step 4: Submit your answer and move to the next task.

Take your time with each passage. The ambiguity may be subtle — focus on which instruction would be hardest for an AI to execute reliably without asking for clarification.

Why It Matters

By contributing to this task, you are:

  • Teaching AI to recognize when instructions lack critical details
  • Building enterprise-grade datasets for safer autonomous systems
  • Improving how AI handles edge cases before taking action
  • Contributing to the sovereign data layer that enterprises and institutions require

This is safety-critical work that AI cannot learn alone — the human layer that makes autonomous systems more reliable.

Task Details

  • Points per question: 100
  • Number of tasks: 10
  • Total points available: 1,000
  • Project eligibility: 1,000 points

Visit the app today to start earning points!

Ready to Get Started?

The Ambiguous Instruction Identification Quest is now live on your Perle Labs dashboard: http://app.perle.xyz

Each identification you submit helps build auditable AI infrastructure for systems where acting on unclear instructions isn't an option — not black-box pipelines, but human-verified data with a clear chain of custody.

Stay Connected

Whether you’re returning from beta or joining for the first time, Season 1 is open and ready. Follow along with us on X, Discord, and Telegram to stay connected so you don’t miss what’s coming. 

Human-verified. Auditable. Sovereign. Start contributing.

New Task Drop: Teach AI to Spot Unclear Instructions Before Acting On Them