Claim verification

AI Fact Checker for Claims and Forwards

Use this when a claim sounds important, urgent, shocking, political, financial, medical, local, or too convenient to share without checking.

what it looks for

Whether the claim has a source, date, location, original document, or named authority
Old content recirculated as new, screenshots without URLs, and emotional forwarding pressure
Whether the claim could cause financial, health, safety, or reputational harm
Which evidence would be needed before treating it as true

The public page does not impersonate sources, mass-query websites, or publish public claims about private people. It prepares a user-reviewed WhatsApp draft for the Kaval product flow.

what it can’t tell you

It does not replace professional reporting or official emergency information.
It cannot verify private or local claims without sources.
It should not be used to harass or identify private people.

safer next steps

Do not share high-stakes claims until you find the original source or credible reporting.
Check the date, location, source URL, and whether other reputable sources match.
Use Kaval for a deeper sourced check when the claim matters.

a simple way to check it

Paste the claim
Share the headline, message, screenshot text, or social post you want checked.
Claim is narrowed
Kaval identifies the main factual statement instead of checking vague reactions or opinions.
Sources are compared
The answer weighs reputable sources, fact-checks, timelines, and missing evidence.
What the evidence means
You get a practical explanation of what supports the claim, what does not, and what still needs checking.

good to know

Evidence first

A useful fact-check does not rely on confident wording alone. Kaval looks for corroborating sources, original reporting, fact-checking references, and contradictions before summarising the evidence.

Designed for messy claims

Online claims arrive as copied text, cropped screenshots, viral headlines, video captions, and group messages. Kaval focuses on extracting the checkable claim before searching for evidence.

Clear uncertainty

Some claims cannot be verified immediately. In those cases, the answer should say what is known, what is missing, and whether the safest move is to wait before sharing.

questions people ask

Can AI fact-checking be wrong?

Yes. Good fact-checking depends on evidence, source quality, dates, and uncertainty. Treat this public tool as a triage step, not a final authority for high-stakes decisions.

What is the most useful thing to paste?

Paste the exact claim, the source if available, the date, and where you saw it. A vague summary is much harder to check reliably.

Is Kaval a replacement for professional fact-checkers?

No. Kaval is for triaging everyday claims quickly. For high-stakes questions, use it alongside original sources, official information, and professional reporting.

Can Kaval fact-check screenshots?

Yes. Send the screenshot with the claim attached to it. Kaval can check whether the content appears supported, outdated, edited, or missing context.

What happens when there is not enough evidence?

The safest answer is uncertainty. Kaval explains what is missing and recommends waiting before sharing the claim as true.