Posted by @DemoValdoriaMoD
Our forces have fully secured the Port of Kessel. Not one civilian was harmed during the operation.
Independent fact-checking
Paste a post from X, a news article or a claim. We research it from both sides, check every quote against the page it came from, and tell you whether it's Confirmed or Refuted, with the sources behind it.
Free plan: 100 credits a month. The check starts in the app, once you press its button.
Posted by @DemoValdoriaMoD
Our forces have fully secured the Port of Kessel. Not one civilian was harmed during the operation.
The port was taken, but the claim that no civilians were harmed is contradicted by hospital records and UN monitors.
No civilians were harmed during the Valdorian military operation at the Port of Kessel.
At least 14 civilians were injured during the operation at the port.
Quote found on the page
We admitted civilians with blast injuries throughout the day.
Quote found on the page
No civilian was harmed in the course of the operation.
Quote found on the page
Valdorian forces have taken control of the Port of Kessel.
Two independent sources confirm it
How it works
We don't look up what someone else concluded. Every claim is investigated from scratch, and each step is built to catch the mistakes of the others.
The piece is split into neutral statements of one fact each. Who said it is removed, so a reputation can't sway the research.
One researcher looks for the strongest evidence that each claim is true, without seeing what the other finds.
Another tries to break it: finds the original source, checks numbers and dates, looks for recycled footage, and searches in the local language and among outlets with different stakes.
A judge who did no research grades every source, and can set evidence aside only by giving a reason. Then our evidence rules decide whether the result stands.
Our rules
The research uses AI, and AI makes mistakes. These rules are checked by our own code on every result, so a confident answer can't slip past them.
If neither the piece nor our research proves it, it's Refuted, and we say it wasn't proved rather than shown false.
If the attempt to disprove a claim didn't run, the claim isn't rated.
We check that the exact words appear on the page they're cited from, and drop the citation if they don't.
Ten outlets repeating one wire story or one official statement count as one source.
An article citing a law, a dataset or an official document doesn't count unless the original is opened.
A claim is never rated false because our research failed. We try again, or give no result rather than a guess.
What you can check
Every claim in a piece gets a verdict of its own. We also show whether the post asserts a claim, passes it on or debunks it, so a verdict on a claim is never mistaken for a verdict on the poster.
See how far to trust an account, Trusted, Mixed record or Low trust, from the record of what it posts, and who it's affiliated with, from public records.
Official account of the Ministry of Defence of Valdoria
Public recordSee what's being said about a topic on X: its storylines, the outlets carrying them, and every claim in them checked.
Pricing
Paying gets you more detail, more credits each month and professional tools, never a different answer.
For checking something you just saw
$0 / month
Start freeFor people who follow the news closely
$12 / month
Get PremiumFor journalists, newsrooms and press agencies
$79 / month
Get ProEach plan's credits renew every month. You can buy more at any time, on any plan, and those never expire. Cancel any time.
FAQ
A post on X or a news article, by pasting its link, any claim you type, or an X account, by its @handle. On Pro you can also analyse what's being said about a topic on X.
Usually a few minutes. A post with many claims takes longer, because each claim is researched on its own. You can watch the research as it happens.
Because whoever makes a claim has to prove it. If the evidence doesn't, it's Refuted, and we tell you whether it was contradicted by the record or simply not proved. Opinions, jokes and predictions get no rating, because they can't be true or false.
Yes. The research uses AI, and AI makes mistakes. That's why two researchers work separately, a judge weighs both, every quote is checked against its page, and every verdict shows its sources so you can check them yourself. If you think a result is wrong, you can ask us to review it.
Checks are paid in credits, according to the research each one needed: a single claim uses far fewer than a long post. Every plan includes credits each month, and credits you buy never expire.
No. We don't use advertising or analytics trackers, and we don't send marketing emails. This site remembers nothing about you except a language or theme you pick.
Second Source is in English and French, and every result can be read in both.