Check news claims before they move.
FactSentinel helps journalists, editors, researchers, educators, and careful readers review article claims with sources, confidence, reasoning, caveats, and model agreement in one visible trail.
News verification starts with the exact sentence.
A headline, quote, statistic, date, or attributed claim can change how readers understand a story. FactSentinel keeps the claim, evidence trail, uncertainty, and model disagreement visible so a human reviewer can decide what needs manual checking.
Highlight the claim
Use the web checker for pasted text or the Chrome extension to check selected article text without losing the context around it.
Review source support
Inspect whether the linked or named sources support the exact wording, not just the general topic.
Flag uncertainty
Treat caveats, low confidence, or model disagreement as prompts for a slower manual pass before sharing or publishing.
Built for news-adjacent verification work.
Use the page when speed matters, but the final decision still belongs to a person who can inspect the evidence and context.
Reporting and editing
- Article claims and backgrounders
- Numbers, dates, and attributed statements
- Draft copy that includes AI-assisted phrasing
Media literacy
- Examples for classroom discussion
- Claims from newsletters or social posts
- Source-trail practice for students
Research review
- Claims summarized from reports
- AI-generated research notes
- Policy or institutional references
Use AI to triage the review, not to skip the review.
FactSentinel is designed to expose signals that guide a human next step: verdict, source links, confidence, reasoning, caveats, and model agreement. It does not replace newsroom standards, corrections policies, or editorial judgment.
Need the Chrome workflow?
Use the Chrome guide when you want to check selected text from the page you are already reading and keep the source trail visible.
Need an automated article triage workflow?
Use the automatic verification guide when a full article has multiple claims and you need a first-pass source trail before deciding what deserves manual review.
Need to verify article sources?
Use the source checker when a story depends on a cited report, agency release, public record, or source page that needs a closer evidence pass.
Need to inspect citations?
Use the citation checker when the article or AI-assisted draft includes formal references, official-looking titles, journal names, reports, or policy documents.
Need a real fake-source case?
The South Africa AI policy case study shows how plausible-looking references can fail under source-level review.
Common questions
What is an AI news fact checker?
It is a first-pass workflow for checking specific article claims with source links, confidence, reasoning, caveats, and model agreement before the claim is trusted or shared.
Does FactSentinel replace reporting or editing?
No. FactSentinel helps expose the review trail. A human reviewer still decides what needs manual verification, correction, context, or escalation.
Should I check a whole article at once?
Start with the claims that carry the most evidence burden: numbers, quotes, dates, named sources, policy references, medical context, legal context, or anything likely to be repeated.
When should I use the Chrome extension?
Use the Chrome extension when you want to check selected text from a live article, source page, research note, or draft without switching out of the browser workflow.
Make the claim trail visible before the story travels.
Use FactSentinel when a news claim deserves a faster first pass, but still needs human review before it becomes trusted context.