Affiliate Disclaimer: This site contains affiliate links — if you purchase through them, we may earn a small commission at no extra cost to you. This helps us keep the platform running and our tools free. Our opinions are always honest and entirely our own. Thank you for your support.
Advertisement
Verify Your Claims with Local AI Fact-Checking
Our free Document Fact-Checker is a revolutionary tool for journalists, researchers, and students. By leveraging advanced Document Question Answering (QA) and Sentence Similarity models, this tool audits your drafts against primary source PDFs. The best part? Your sensitive documents never leave your computer. Everything is processed locally in your browser for absolute privacy and zero server latency.
How to Fact-Check Your Document
- Step 1: Initialize the Engine. Wait for the local AI models to load into your browser. This creates a secure, private environment for analysis.
- Step 2: Paste Your Draft. Enter the text or claims you want to verify in the editor on the left.
- Step 3: Provide the Evidence. Upload the source PDF document that contains the “ground truth” data.
- Step 4: Run the Audit. Click “Verify Claims.” The AI will break your draft into sentences and search the PDF for supporting evidence.
- Step 5: Review the Report. Instantly see which claims match the source text and which ones may need further revision.
Frequently Asked Questions
Is my document secure during the fact-checking process?
Yes. This tool uses “Local Scoping.” Your PDF and your draft are processed entirely within your browser’s memory. No data is ever uploaded to a server or stored in the cloud.
What is “Document Question Answering”?
It is a specialized AI technique where the model reads a document and answers specific questions about it. We use it to “ask” the PDF if the claims in your draft are supported by the text.
Why does it take a moment to load initially?
To ensure your privacy, the tool must download the AI models (DistilBERT and MiniLM) directly to your browser’s local cache. This only happens once, and subsequent uses will be much faster.
What happens if the AI can’t find a match?
The claim will be flagged with “No Direct Match.” This doesn’t necessarily mean the claim is false, but rather that the specific wording or data wasn’t semantically identical to information found in the PDF provided.
Does it support languages other than English?
The current model is optimized for English text. While it may provide results for other languages, the accuracy of the semantic similarity check is highest with English documentation.
Is there a file size limit for the PDFs?
The limit is based on your device’s available RAM. For the best performance and accuracy, we recommend using documents under 50 pages (roughly 15,000 words) for the verification context.