Principle
Belief is explicit
Kalpae expresses uncertainty as a transparent probability state with visible evidence context.
Learn Kalpae
A clear model of how probabilities are formed, revised, and recorded.
Last updated
Feb 9, 2026, 02:12 GMT+5:30
How Kalpae works
Kalpae is built for careful forecasting under uncertainty. This page explains the mechanics, boundaries, and trust model in practical terms.
Principle
Reasoning is inspectable
Every meaningful movement is paired with a short explanation of what changed and why it mattered.
Principle
Participation is simulated
Users forecast through simulated YES and NO positions to build judgment without financial exposure.
01 Overview
What Kalpae does and does not do
The boundaries below are intentional. They protect interpretability and reduce confusion between forecasting practice and speculative activity.
Kalpae does
- Tracks how uncertainty evolves across real-world questions over time.
- Integrates verifiable external signals and updates probabilities using rule-based logic.
- Lets users express forecasts and compare personal beliefs against system and aggregate views.
- Maintains auditable historical traces for probability shifts and resolution outcomes.
Kalpae does not
- Does not support real-money wagering, payouts, or speculative trading behavior.
- Does not treat AI output as unquestionable truth without traceable evidence context.
- Does not hide core update logic behind opaque, unexplained model decisions.
- Does not optimize for engagement pressure or impulsive decision-making loops.
02 Probability formation
How probabilities are formed
System probability updates follow a fixed chain from signal intake to publication. User disagreement is tracked as a separate layer.
Process diagram
Step 01
Signal intake
Structured public signals are collected from approved sources.
Input quality checks run before downstream processing.
Step 02
CRE execution
Chainlink CRE orchestrates off-chain workflow steps for normalization and evaluation.
Execution traces are retained for reproducibility and review.
Step 03
Belief update
Rule policies convert validated evidence into explicit probability adjustments.
Reason labels explain direction, confidence, and magnitude.
Step 04
User disagreement layer
Aggregate user forecasts are tracked as a separate belief signal, not merged blindly.
System belief and crowd belief remain visually distinct.
Step 05
Publication
Updated probabilities and rationale snapshots are logged with deterministic commitments for transparent history.
On-chain anchoring is rolling out progressively by environment.
Separation rule
System probability and aggregate user forecast are intentionally shown as distinct states so disagreement remains visible and interpretable.
03 User workflow
What users do in Kalpae
Forecasting is iterative. Users form a belief, revisit it as evidence changes, and accumulate a record over time.
Workflow diagram
01
Forecast
Pick questions, review the current probability state, and take a simulated YES or NO position with intent.
02
Revisit
As new evidence arrives, revise your position and observe how your belief differs from system and aggregate forecasts.
03
Build record
At resolution, outcomes are recorded into a long-horizon forecasting record focused on calibration and consistency.
04 Infrastructure rationale
Why blockchain is used
Blockchain is used as an integrity layer for records and verification, not as a speculative product surface.
Immutability
Critical belief transitions and resolution records are being introduced to immutable infrastructure so history cannot be quietly rewritten.
Transparency
Publicly verifiable records support external scrutiny of what changed, when it changed, and what logic path was applied.
Verifiability
Users and teams can independently confirm that published states match recorded execution outputs.
05 Common misunderstandings
Clarifications
These are the three most common points of confusion and the direct platform answers.
QIs this gambling?
No. Kalpae does not involve real-money stakes, payouts, or wagering mechanics. Forecasting is simulation-based and used for decision training and performance review.
QCan probabilities be manipulated?
Kalpae separates source-driven system belief from user forecasts and records update paths with timestamps. This makes unauthorized or unexplained changes detectable.
QIs AI deciding truth?
No. Kalpae models uncertainty, not truth. AI-assisted components may help process inputs, but final updates are governed by explicit rules and evidence review.