Don’t just trust an AI answer.
Check the evidence.
AI can sound confident even when a claim is wrong. ClaimGuard helps you examine factual statements using retrieved sources and AI-assisted analysis, so you can better understand what the available evidence does — and doesn’t — support.
ClaimGuard is a free experimental AI-assisted service. Its reports are not independently reviewed and may contain mistakes or incomplete evidence. Do not submit passwords, confidential files, or sensitive personal information. Verify important claims using original sources.
How ClaimGuard works
Three steps take you from a factual statement to an evidence-based, AI-assisted report.
Submit your claim
Enter a specific factual statement and the email address where you would like to receive the report. No Google Forms or separate sign-up page.
Review retrieved evidence
The automated workflow searches for relevant web sources. An AI model examines the retrieved information, compares claims, and identifies gaps or contradictions.
Receive your report
ClaimGuard emails an AI-generated verdict, an explanation, source information, and limitations for you to evaluate independently.
What you should know
ClaimGuard is designed to encourage careful verification, not to replace human judgment.
AI-assisted, not independently verified
Reports are generated automatically. A human editor does not review each verdict before it is emailed. A verdict is not a guarantee of factual accuracy.
Retrieved sources may be incomplete, outdated, misleading, or unrelated to important parts of a claim. AI-generated explanations and citations can also contain mistakes. Always open and assess the original sources yourself.
Submitted claims and email addresses are processed through third-party services used by this workflow, including Make, Tavily, Google Gemini, Google Sheets, and Gmail. Information may be stored as part of that processing.
Do not submit passwords, financial account numbers, medical records, personal identifiers, confidential documents, or other sensitive information.
Do not rely on this tool alone for medical, legal, financial, regulatory, or other high-stakes decisions.
Frequently asked questions
Is ClaimGuard free?
Yes. ClaimGuard is currently available as a free experimental beta. Usage limits may apply to keep the service available.
How long does verification take?
Reports are processed automatically, but delivery time can vary depending on system availability, queued requests, and the external services used. Check your inbox and spam folder.
Can I trust every verdict?
No. The report is AI-assisted and may be incorrect. Treat each verdict as a starting point for further investigation. Review original sources and verify important details independently.
Where will I receive the report?
The report is sent to the email address submitted with your claim. If you do not see it, check your spam or junk folder. Submission confirmation does not guarantee successful email delivery.
What kinds of claims work best?
Clear, specific, checkable factual statements work best. Claims involving dates, published standards, official policies, documented statistics, or product specifications are usually easier to investigate than broad opinions or predictions.
Does ClaimGuard check live websites?
ClaimGuard uses retrieved search information. It does not guarantee that every linked webpage was opened and independently inspected. You should review source URLs and confirm that each source actually supports the reported claim.
Is my report reviewed by a person?
No. The automated report is not individually reviewed by an editor before delivery. Human verification remains necessary.
ClaimGuard by The Circuit Edit
A free experimental tool for more careful thinking about AI-generated information.
Explore The Circuit Edit →