PLATFORMS
Misinformation Virality Detector (MVD)
Research-grounded detection of new and emerging misinformation, with an assessment of how far it is likely to spread.
The problem
Challenge
Most detection tools match keywords against claims already known to be false, so they find narratives that are already spreading. New narratives move faster than that process, and by the time a claim is confirmed false, many people may already believe it.
What it changes
Impact
The Misinformation Virality Detector is based on research from the University of Zurich, and built from real information challenges, drawing on years of R4P work in social listening, misinformation and communication. It reads how content is built across four structural patterns that manipulative content shares across topics: language, narrative, logic and critical thinking. It then estimates how far the content is likely to travel. It does not check claims against a list of facts. That means it can flag a claim as soon as it appears, on a subject nobody was monitoring.
Under the hood
How it works
Four steps, from raw conversation to a briefing an analyst has already read.
01
Bring in your data
Use it through an API inside your own systems, upload a dataset, run it within our monitoring platform, or check a single page through the browser extension.
02
Risk scores
Content is rated from likely factual to very likely misinformation, with the manipulation patterns that produced the score set out alongside it. You can complement it with your own categorization model built in this tool.
03
Assess the spread
The system estimates how viral the content is likely to become and maps how the narrative is moving across platforms, so the response goes where it will matter.
04
Fit to your team's workflow
Results can be explored with a tailored AI assistant, exported as a report for colleagues who need the summary, or connected to an existing monitoring system.