Truth Seeker is an AI-powered fact-checking engine that analyzes claims, URLs, and social media posts in real time using web search and multiple AI models.
ANALYZE A TEXT CLAIM
1
Type any claim, question, or statement into the Truth Seeker input box on the main dashboard.
2
Press ENTER or click the analyze button.
3
The AI searches the web, cross-references sources, and returns a Credibility Score (1-10), Bias Rating, and detailed analysis with cited sources.
💡You can ask anything — "Is it true that...", "Did [person] say...", or just paste a headline.
ANALYZE A URL
1
Paste any news article, blog post, or social media URL into the input box.
2
A link preview bar appears showing the URL — you can now add optional context like "focus on the inflation claims" before analyzing.
3
Press SEEK TRUTH. The AI fetches the content, extracts claims, and analyzes them.
4
Works with most websites including Twitter/X Reddit Instagram YouTube and news sites.
💡Facebook links: Facebook blocks server-side access. If you see a warning, just copy the text from the Facebook post and paste it directly into Truth Seeker instead of the URL.
FACT-CHECK A YOUTUBE VIDEO
1
Paste any YouTube video URL (youtube.com/watch, youtu.be, Shorts, or embeds).
2
Truth Seeker automatically extracts the full video transcript from captions/subtitles — no manual copying needed.
3
The AI reads the transcript, identifies up to 5 key factual claims, and fact-checks each one individually with cited sources.
4
You get a video summary, per-claim verdict, overall credibility score, and bias rating. The YouTube thumbnail is saved for easy identification in the archive.
💡Most YouTube videos have auto-generated captions. If a video has no captions, Truth Seeker falls back to the title, description, and web search results about the video.
ANALYZE AN IMAGE
1
Paste an image from your clipboard, drag & drop an image file, or click the image icon next to the input to upload.
2
A preview thumbnail appears. Add optional context like "is this headline real?" in the text input.
3
Press SEEK TRUTH. The AI reads all text in the image, identifies claims, checks for manipulation, and verifies facts via web search.
4
Results include an Image Description, Manipulation Check (editing, AI-generation, out-of-context use), plus the standard credibility and bias scores.
💡Images are compressed to 1200px max before sending. Supports JPG, PNG, and most image formats. The analyzed image is saved and displayed on the permalink page.
ONE-CLICK ANALYSIS FROM FEEDS
1
Look for the ⚡ button next to any headline in the news feeds, Reddit Pulse, Hacker News, Trending Searches, or GDELT panels.
2
Click it to instantly send that article to Truth Seeker for analysis.
UNDERSTANDING YOUR RESULTS
Credibility Score — How well does the evidence support this claim? 1 (unsupported) to 10 (well sourced) Evidence Score — How much verifiable evidence was found? 1 (minimal) to 10 (overwhelming from multiple credible sources) FAR LEFTCENTERFAR RIGHT
Bias Rating — Political lean of the source💡High credibility + strong evidence = well-documented. Low credibility + strong evidence = widely disputed by sources. High credibility + limited evidence = plausible but thin sourcing. Always check cited sources yourself for important claims.