Trust Score & Risk Scoring Policy

1. Purpose & Strategic Role

This policy defines the framework for evaluating user trustworthiness and risk levels within Glancers.

The Trust Score system is a risk-and-reputation scoring mechanism that:

Enhances platform security
Improves decision-making (matching, payouts, visibility)
Rewards reliable behavior
Detects and mitigates fraud risks

It transforms user activity into actionable trust signals across the platform.

2. Trust Score Model

2.1 Definition of Trust Score

The Trust Score is a dynamic numerical indicator representing a user’s overall reliability, based on behavioral, financial, and professional performance.

The score may be updated as new activity is recorded.

2.2 Multi-Dimensional Scoring Framework

The Trust Score is calculated across multiple dimensions:

Behavioral Dimension: - Response time - Deadline adherence - Policy compliance

Financial Dimension: - Payment success rate - Dispute frequency - Transaction consistency

Professional Dimension: - Client ratings - Project completion volume - Skill specialization consistency

2.3 Data Weighting & Recency

Recent activity is weighted more heavily than historical performance
High-value transactions carry greater influence
Sustained consistency is rewarded over short-term spikes

This ensures the score reflects current reliability, not legacy performance.

2.4 Algorithm Confidentiality

The detailed scoring logic, weights, and thresholds are strictly confidential
Disclosure is limited to high-level guidance only

To balance transparency and integrity, Glancers provides:

Proactive guidance signals to users (e.g., response delays, dispute frequency)
Actionable recommendations to improve Trust Score without exposing core algorithms

This prevents:

System manipulation
Reverse engineering of ranking signals

3. Risk Scoring & Detection

3.1 High-Risk Signals

The system detects and reacts to suspicious behaviors, including:

Sudden geographic changes
Repeated failed payment attempts
Rapid dispute escalation patterns
Abnormal activity spikes

Context-aware safeguards are applied to reduce false positives, including:

Differentiation between legitimate travel (e.g., digital nomads) and account compromise
Device consistency and behavioral continuity checks

Such signals may trigger:

Trust score reduction
Temporary restrictions

3.2 Network Risk Analysis

User risk is influenced by connected accounts through:

Shared devices or networks
Transaction relationships
Behavioral similarity patterns

An association with a high-risk account is not, by itself, a basis for reducing a User's Trust Score; network signals may prompt a review, and any score impact requires that review to find the User's own involvement.

3.3 Identity Verification Impact (KYC/KYB)

Verified users receive higher baseline trust levels
Additional verification increases system confidence

Unverified or partially verified accounts may:

Face limitations
Be subject to stricter monitoring

3.4 Behavioral Anomaly Detection

Machine learning models monitor:

Deviations from normal usage patterns
Non-human interaction signals
Automated or scripted behaviors

Detected anomalies may result in:

Score penalties
Security reviews

3.5 Human Review & Appeal Mechanism

In cases where Trust Score impacts critical user rights (e.g., account restriction, restriction affecting compensation processing or payout processing):

Users may request human review
Dedicated risk teams will evaluate edge cases
Final decisions may override automated outputs when justified

This ensures fairness and safeguards against AI misclassification

4. Trust Score Impact on Platform Privileges

Trust Score directly influences user capabilities, including:

Visibility in search and ranking
Access to high-value projects
Payout-processing speed, review requirements, and applicable operational limits
Bidding capacity
Eligibility for premium features

Higher scores unlock enhanced privileges.

Lower scores may result in restrictions.

4.1 Trust Recovery & Rehabilitation

Glancers supports structured recovery of Trust Score through:

Consistent positive behavior over time
Reduced dispute activity
Improved responsiveness and compliance

The system applies:

Gradual score restoration mechanisms
Decay of past negative signals when sustained improvement is detected

This ensures users are not permanently penalized for recoverable behavior

5. Fairness, Bias & Safeguards

Glancers is committed to ensuring scoring fairness:

Periodic review of scoring models, where implemented
Monitoring for geographic or linguistic bias
Prevention of unfair penalization

The system is designed to reflect behavior, not identity.

6. Governance Principle

Trust Score is treated as a quantifiable, behavior-based metric within Glancers, not a financial asset or a form of property belonging to the User.

Earned through consistent behavior
Lost through risk signals and violations
Continuously recalculated

The Trust Score system ensures that:

Honest users are rewarded
Risky behavior is contained
Platform integrity is preserved

Procedural Review & Appeals

Any adverse administrative action, temporary restriction, or account decision taken under this policy is subject to procedural review and may be appealed in accordance with the Appeals & Reconsideration Policy.

Version History

2026-09 — Qualified descriptions of Platform features and service targets. Clarified enforcement, temporary restrictions, and appeal rights. Editorial and formatting corrections.
2026-09 — Revised selected provisions for accuracy and consistency with related policies.
2026-09 — Clarified consistency with related policies and qualified statements about Platform features.
2026-09 — Updated payment, compensation, and relationship terminology to reflect that Clients purchase Services from Glancers and that Glancers separately engages and compensates Freelancers.