Methodology

How TruthLens works

TruthLens uses a multi-step, evidence-first approach. We never make assumptions — every verdict is grounded in retrievable evidence, and uncertainty is always disclosed.

Core principles

Strict neutrality

No political, cultural, or ideological lean. Evidence only.

Full transparency

Every verdict shows its reasoning, sources, and confidence level.

Source-first analysis

Claims are matched against real-world news and official sources.

AI + human logic

Smart reasoning engine analyzes evidence using journalism-inspired prompts.

Uncertainty is honest

When evidence is insufficient, we say Unverified — not guess.

No assumptions

The AI is instructed never to assume beyond what evidence shows.

Verdict levels explained

Verified

Strong evidence from trusted sources confirms the claim.

80–100
Likely True

Evidence leans toward truth but is not fully conclusive.

65–79
Mixed / Context Needed

Some aspects are true, others false, or important context is missing.

40–64
Misleading

Technically true but framed in a way that misleads.

30–49
Unverified

Insufficient evidence available to make a determination.

Likely False

Evidence leans against the claim being true.

15–29
Disputed

Credible sources directly contradict each other.

0–29

Credibility score weights

Trusted source confirmations30%
Evidence quality and quantity25%
Contradicting sources20%
Domain reputation signals15%
Language sensationalism10%

Important disclaimer

TruthLens is an AI-powered tool designed to assist research. Results are not infallible. For high-stakes decisions, always consult primary sources, official publications, and qualified experts.