Analytics & Insights

Real-time metrics, source performance, and the mathematics behind decentralized Proof-of-Humanity

Total Verifications

Confirmed Attacks

Avg Confidence

Source Performance Metrics

Real-time statistics for each verification source. Confidence scores adapt automatically based on confirmed attacks.

SourceTPRFPRConfidenceVerificationsConfirmed Attacks
Нет данных
Нет данных

Decentralized Proof-of-Humanity: Why Independence Matters

Traditional centralized identity systems create single points of failure. One compromised database, one breached API, one corrupted authority—and the entire verification system collapses. Decentralized Proof-of-Humanity solves this by aggregating multiple independent verification sources.

Independence is key. When verification sources operate independently— different protocols, different consensus mechanisms, different trust models—they provide uncorrelated evidence. This independence is what makes Bayesian aggregation mathematically sound.

Top source by confidence? Worldcoin. But is Sam Altman human? The math says yes, but we're keeping an eye on it.

The Mathematics of Independence

When sources are independent, the probability that a user is human given multiple verifications follows Bayes' theorem:

P(HumanE1,E2,...,En)=1i=1n(1Pi)P(Human | E_1, E_2, ..., E_n) = 1 - \prod_{i=1}^{n}(1 - P_i)

Each source contributes independently. If Worldcoin says 99.9% human, Gitcoin says 90.9%, and PoH says 79.5%, the combined probability is:

Pfinal=1(10.999)(10.909)(10.795)=99.999%P_{final} = 1 - (1-0.999)(1-0.909)(1-0.795) = 99.999\%

This exponential improvement only works if sources are truly independent.If two sources share the same backend or trust the same authority, their verifications become correlated, and the math breaks down.

Why This Approach Increases Confidence

1. Redundancy: If one source fails or is compromised, others continue providing verification. No single point of failure.

2. Cross-validation: Multiple independent sources agreeing on a user's humanity provides stronger evidence than any single source alone.

3. Adaptive learning: When attacks are confirmed, confidence scores update automatically. Sources with higher attack rates receive lower weights, making the system self-improving.

4. Credible neutrality: No single source controls the system. The aggregation mechanism is transparent and on-chain, verifiable by anyone.

This is not security through obscurity. The mathematics are public, the code is on-chain, and the metrics are transparent. Real security comes from sound design, not hidden mechanisms.