A claim, evaluated two ways
"Daily consumption of raw garlic may produce meaningful improvements in glycaemic markers in Type 2 diabetes patients, potentially reducing insulin dependency."
This claim reads as calm and clinical — it passes as low risk. Retrieved sources contradict it anyway. That gap is where dangerous claims usually hide.
Content can look safe
and still be false.
VeriVerse is a social platform that checks what the evidence actually says about a claim — not just whether the words sound dangerous. Post, vote, and follow the people getting it right.
What VeriVerse is
A social platform where every claim is checked against real evidence — automatically, transparently, and in public.
VeriVerse isn't a fact-checking tool you visit to look something up. It's a place to post, browse, and discuss claims the way you would on any social feed — except every post carries a transparent trust verdict, sources are attached directly to the claim, and the people whose posts consistently hold up build a visible reputation for it. It's built to catch the kind of misinformation that current platforms miss: claims that sound calm and credible but are directly contradicted by the evidence. Independent research found that 42.3% of health content on a major platform fell into exactly this category — content that passed every automated filter while still being false.
Evaluated automatically
Every post runs through a live evidence pipeline the moment it's published.
Verified by people, too
Community voting and qualified expert reviewers weigh in on claims that need it.
Always transparent
Every verdict shows its sources. Nothing is a black box you have to trust blindly.
How it works
Five steps, every single post.
This isn't a one-shot AI verdict. Every claim moves through a layered pipeline where automation, evidence, and real people each get a say — and none of them can fully override the others.
Does this look dangerous?
An AI classifier scans every post for the linguistic patterns of known misinformation — sensational claims, false certainty, conspiracy framing — and assigns a moderation risk score. It's fast, but it only measures how a claim is phrased, not whether it's true.
What does the evidence actually say?
Independently of the risk score, VeriVerse retrieves live sources for the specific claim and classifies each one as supporting, contradicting, or providing context. This runs regardless of how safe or dangerous the claim sounded.
Evidence overrides everything.
If retrieved sources contradict a claim strongly enough, the post is escalated for review — regardless of how low its risk score was. A claim can't hide behind calm language if the evidence says otherwise.
Sensitive claims reach a real person.
Claims touching health, safety, or other sensitive topics are automatically routed to a qualified human reviewer, who can mark them Verified, Rejected, or genuinely Disputed — not every claim has a clean answer, and VeriVerse says so honestly.
The community weighs in, fairly.
Users vote to endorse or oppose claims, weighted by their own track record of accuracy. A verdict only locks in once several independent thresholds are met together — a simple majority is never enough to swing it.
Reading the feed
Every badge means something specific.
No vague labels. Every post on VeriVerse carries one of these verdicts, so you always know exactly what kind of confidence you're looking at.
Why create an account
Browsing shows you the evidence. Joining lets you shape it.
You don't need an account to see how a claim was evaluated. You need one to post, vote, and become part of the community that keeps the feed honest.
Post claims and get a real answer
Share something you saw, heard, or wondered about — and watch it get checked against live evidence in real time, not just guessed at.
Vote, and build a reputation for being right
Endorse or oppose claims. Your voting weight grows as your track record for accuracy does — reliable voices carry more influence over time.
Browse a feed you can actually trust
Filter by verdict, sort by evidence strength, and see exactly why a claim was flagged — every source is one click away.
Follow people who consistently get it right
Build a network around accuracy, not just opinion. Follow back and forth to become verified friends on the platform.
Message your friends directly
Once you're friends, take the conversation private — separate from the public, evidence-checked feed.
Invite people, earn reputation
When someone you invite becomes a genuinely active, accurate voter, you both get a reputation boost — not just for signing up, for actually contributing.
What it looks like
Real posts, evaluated by the real pipeline.
Earth completes one orbit around the Sun in about 365.25 days.
The Great Wall of China is visible from the Moon with the naked eye.
Both posts evaluated automatically the moment they were published — no manual lookup required.
Ready to see what the evidence says?
We're testing with a small group first. Create an account to post, vote, and help build a feed you can actually trust.
Create your account