Methodology
AfghanVerified aggregates news from Afghan and international outlets and assigns a reliability label to each story. This page explains what those labels mean, how they're assigned, and where the limits are. Each story also carries an editorial tone assessment that evaluates the language quality of the original sources.
Reliability labels
Every story on AfghanVerified carries one of three labels:
Reliable
Multiple credible outlets are reporting the same story, or the report comes from a well-established source with a strong editorial track record. This doesn't mean every detail is confirmed, but the sourcing is solid enough to take seriously.
Unconfirmed
The story is being reported but coverage is limited, details are still emerging, or the available sources have mixed track records. Worth following, but treat specifics with caution.
Unverified
Single-source reporting, unclear sourcing, or the claim hasn't been picked up by other outlets. This doesn't mean it's false, just that there isn't enough independent reporting yet to assess it.
Editorial tone
Separately from reliability, we evaluate the rhetorical quality of the original source texts. This measures how outlets communicate, not what they believe. The article you read on AfghanVerified is always rewritten in neutral language. The tone label tells you about the sources it was built from.
Neutral
The source texts use neutral, wire-service style language. Claims are attributed, facts are presented without emotional loading, and no opinion language or hyperbole is present.
Framed
The sources mix factual reporting with opinion language, mild emotional framing, or advocacy phrasing. Examples include loaded adjectives like "brutal crackdown" or "desperate plea," or unnamed value judgments presented alongside facts.
Sensationalized
The sources use heavy emotional manipulation, hyperbole, or present opinion as established fact. This includes extreme superlatives, clickbait language, conspiracy framing, or dehumanizing language toward any group.
What tone does not measure: Political terminology choices (such as “Taliban” versus “Islamic Emirate”) are not flagged. These are political labels, not rhetorical techniques. Direct quotes from officials using strong language are also not flagged if the article itself reports them neutrally. Editorial selection (which facts a source chose to cover) is inherent to journalism and is not penalized.
How labels are assigned
Reliability and tone are assessed independently by AI. Reliability evaluates sourcing quality: how many independent outlets are covering the story, whether claims are attributed to named sources, and the overall consistency across reports. Tone evaluates language quality: whether the original sources use neutral reporting or loaded, emotional language.
AI-generated labels are then subject to human review. Our team spot-checks assessments, corrects mislabels, and can override any rating. The goal is to combine the speed of automated processing with the judgement that only a person can provide.
Each article includes a short reasoning line that explains why a particular reliability label was assigned, and when the tone is framed or sensationalized, specific loaded phrases are cited from the source texts. Both are visible on every article page.
Sourcing and citations
We don't produce original reporting. Every story links directly to the primary source article, and when multiple outlets cover the same event, we list them all under “Also reporting.” You can always click through to read the original in full.
Our source list currently includes Afghan outlets (ToloNews, Ariana News, Pajhwok, Hasht-e Subh, Khaama Press, Amu TV, Afghanistan International) and international outlets (BBC Persian, Al Jazeera, AP News). We plan to expand this over time.
Limitations
Reliability labels are an indicator, not a guarantee. They reflect source-level patterns and cross-referencing — they are not a verdict on whether any specific claim is true or false.
We cover a curated set of outlets, not the entire Afghan media landscape. Stories that only appear in sources we don't track won't show up here. Labels are AI-assisted with human review, but may occasionally misjudge — if you spot something that looks off, please let us know.