How we use AI

WeNepal is written by a newsroom team that uses automated tools, including AI, for some of the work. This page explains in plain language where stories come from, who does what, what the software is not allowed to do, and what you will see on a story.

• Our staff report, write, edit, approve and post stories. Software helps them find, sort and draft.
• We build stories from public reports and official information and rewrite them in our own words. We do not name other news outlets.
• Every fact in a draft is checked against the sources before anyone sees it.
• Some stories are prepared automatically and publish after passing every check. They carry the byline "LacNova AI", the name of WeNepal's own newsroom AI, so you can always tell. Editors check a sample of them every day.
• Stories on sensitive subjects (elections, courts, deaths, communal issues, named individuals, minors and health emergencies) publish only when they also pass strict extra checks. A story that fails is fixed once automatically; if it still fails, it is not published.
• We keep our sources on record for review and corrections, and when we get something wrong, we correct it in public.

Our newsroom follows public reporting from Nepali and international media and primary sources such as Nepal Rastra Bank, NEPSE, the Election Commission, the Supreme Court, the Department of Hydrology and Meteorology, the Nepal Gazette and ministry press pages. Each source has a trust score that goes down if it is repeatedly wrong and up if it holds.

The software only reads what sources publish. It does not go looking on social media, and a social post counts as a source only when the account is verified as the person or institution.

A story starts in one of two ways. A staff member builds it, or, when two or more independent reports or one primary source cover the same event, our tools prepare a draft: a summary of 80 to 120 words, three bullet points and a "why it matters" line. The Nepali and English editions are written separately, in each language's natural order, rather than translated one from the other.

The writing rules are strict:

1. Use only facts found in the sources.
2. Every number, name, date and quote must appear in at least one source.
3. Write in our own words; never copy a sentence, a headline, a photo or a video.
4. Keep every source with the story, on record for our editors.

Every draft goes through a set of checks before a reader sees it. A draft that fails a check is fixed once automatically; if it still fails, it is not published.

Staff build, edit and approve stories, and some stories are prepared and published automatically when they pass every check.

Stories on sensitive subjects (elections, court cases, deaths and violence, communal issues, named individuals, especially those facing allegations, health emergencies and anything involving minors) publish automatically only when they also pass strict checks:

1. At least two independent sources, or an official primary source such as a gazette notice, a court order or an official press release.
2. Neutral attribution ("police said…").
3. No accusation stated as fact.
4. No graphic detail, and nothing sub judice.
5. Minors are never named.

A story that fails is fixed once automatically. If it still fails, it is not published.

Editors review a sample of automatic stories every day and can edit, correct or unpublish any story at any time. Corrections follow our corrections policy.

Any editor or engineer can pause the entire newsroom at any time.

If a story is wrong, we correct it and say so on the story, dated. Major corrections are also posted on the channels where the story appeared and listed on the public corrections page. A story that should not have been published is retracted, not quietly deleted. Because we keep every story's sources on record, editors can trace where an error came from; sources that turn out to be wrong lose trust, and sources that are wrong repeatedly are dropped.

Some of our videos use a synthetic voice. Where we publish them on YouTube, Facebook, Instagram or TikTok, we turn on the platform's own altered or synthetic content setting.

• Invent a quote, a number, a name or a date.
• Publish a story on a sensitive subject that has not passed the strict checks.
• Generate an image of a real person, or clone a real person's voice.
• Copy another outlet's paragraphs, headlines, photos or video.
• Publish a fully automatic story without the "LacNova AI" byline that says it was prepared automatically.

Write to hello@wenepal.com. To report an error, use corrections@wenepal.com or the contact form. Our full rules are in the editorial policy.