A false claim does not need a translator. It can be copied, paraphrased and adapted into a new language within hours, often losing whatever caveats it once had. Corrections move more slowly. A fact-check published in one language may never reach the readers who saw the claim in another.
The gap between languages
Most fact-checking resources, and most of the data AI models are trained on, are in English. Readers of other languages often have fewer independent fact-checking outlets and fewer tools. That gap is exactly where misinformation finds room to grow.
Why English, Russian and Persian
Honest News works in three languages with very different characteristics:
- English, the language of much international reporting and research;
- Russian, with a large audience across many countries and its own rich media landscape;
- Persian, written right to left, with its own script, numerals and typographic conventions.
Supporting all three from the beginning forces the platform to take language seriously, rather than translating an English product at the end.
What multilingual fact-checking requires
- Answers in the reader's language. A claim checked in Persian is explained in Persian, even if the sources are in English.
- Honest typography. Persian text needs right-to-left layout, its own fonts, and no letter-spacing that breaks joined letters. Mixed text, such as a Persian sentence with an English source name, must display correctly in both directions.
- Cross-language evidence. A claim in one language can be checked against sources in another.
- Awareness of translation drift. A mistranslated word can change a meaning, and should be treated as a possible source of distortion.
How to use it
In Honest News, choose EN, RU or FA at the top of the page. The whole interface changes, including layout direction, and the analysis is written in the language you choose. Shared report links open in the language of the report.
Truth should not depend on which language you happen to read.