6 Proven Verification Techniques
Fact-checkers use these methods every day. Most take a few minutes at most, and together they catch a lot of what goes viral.
Reverse Image Search
Before trusting a viral image, find out where it came from. Search it with Google Lens / Google Images or TinEye to find earlier copies and the original source. Many viral images are repurposed from unrelated events or generated entirely by AI. (BSmeter does not detect AI-generated or manipulated images — this is a check you do yourself.)
- Take a screenshot of the image or a video keyframe you want to verify
- Upload to Google Images (images.google.com) or TinEye (tineye.com)
- Check if the image appears earlier with different captions or dates
- Look for Content Credentials (C2PA) or a platform AI label — visual glitches like odd fingers are less reliable than they used to be
Lateral Reading
The single most effective verification technique used by professional fact-checkers. Instead of staying on one website and evaluating it in isolation, open new tabs and search for what other sources say about the same claim or organization. The credibility of a source is determined not by its own website design, but by what other trusted sources say about it.
- Open a new tab before you finish reading the current page
- Search for the author, organization, or claim mentioned
- Check what established sources (Reuters, AP, academic journals) say
- If no credible source covers the story, treat it with extreme skepticism
Source Tracing
Every claim in a video or article should be traceable to an original source. Misinformation often cites other sources that themselves are recycling unverified claims. The technique is simple: follow the citation chain all the way back. If a YouTube video cites a news article that cites a tweet that cites 'someone said' — the chain is broken.
- Note every specific claim: statistics, dates, named studies
- Click through every link in the description or article
- If a study is cited, search for the actual paper title
- If the original source doesn't support the claim, the claim is misleading
Check the Checker
Trusted fact-checking organizations like Snopes, PolitiFact, FactCheck.org, and Reuters Fact Check have already investigated thousands of claims. Before spending time verifying something yourself, search these databases. If the claim has already been debunked by multiple independent fact-checkers, you have your answer instantly.
- Search Snopes.com for the specific claim or keywords
- Check PolitiFact for political claims and statements
- Search Reuters Fact Check (reuters.com/fact-check)
- For videos and podcasts, use BSmeter: paste the link for a claim-by-claim analysis with sources on the weakest claim (the 3 weakest on Basic and Premium)
Context Verification
Out-of-context clips are one of the most common forms of video misinformation. A 5-second clip from a 2-hour speech can completely reverse the speaker's meaning. Always seek the full context before accepting a viral clip as evidence. BSmeter shows the transcript and can flag a claim as Missing Context, but it only sees the clip you paste — it can't recover what was cut.
- Search for the full video, speech transcript, or article
- Look for when the clip was recorded — old footage is often passed off as current
- Check who recorded/created the clip and why
- Verify dates: misinformation often resurfaces years-old footage with updated captions
SIFT Method
Developed by digital literacy researcher Mike Caulfield, the SIFT method is a four-step framework for evaluating online information: Stop, Investigate the source, Find better coverage, and Trace claims to the original context. It is designed to be quick enough to use on everyday social media content.
- S — Stop: Do not share, like, or react until you have verified
- I — Investigate the Source: Who created this and what is their reputation?
- F — Find Better Coverage: Is a trusted news outlet reporting the same thing?
- T — Trace to Original: Find the original claim, quote, or video and verify it directly
Red Flags: When to Be Suspicious
These warning signs indicate content that needs extra scrutiny before you share or act on it.
Urgency & Emotional Manipulation
Content that demands immediate action ('SHARE BEFORE THIS IS DELETED', 'You won't believe what happens next') is designed to bypass your critical thinking. Real news doesn't need emotional manipulation.
What to do: Close the tab. Search for the claim from an established news source.
No Author or Anonymous Sources
Credible content is attributed to named individuals or organizations. 'A source says' or 'Experts claim' without specifics is a major red flag. Anonymous posts on social media are the least verifiable content type.
What to do: Search for the author's name. If no verifiable author exists, the content is untrustworthy.
Missing or Broken Citations
Legitimate journalism and research provide links to primary sources. If a video or article makes claims without linking to studies, official documents, or named sources, those claims cannot be verified. Treat them as unsubstantiated.
What to do: If no citation exists for a factual claim, assume it is unsupported until proven otherwise.
Perfection — Possible AI-Generated Content
AI-generated content is increasingly realistic. Sometimes there are signs — overly smooth skin, lighting that doesn't cast believable shadows, text that garbles on close inspection, unnatural eye reflections — but newer tools often leave none. BSmeter does not detect AI-generated media, so don't rely on any single visual tell.
What to do: Look for Content Credentials (C2PA) or the platform's AI label, reverse-search the image or a keyframe, and trace it to the original upload.
Master the SIFT Method
Created by digital literacy researcher Mike Caulfield, SIFT is the gold standard for quick online verification.
S. Stop
Before you share, like, comment, or even fully read — stop. Your first instinct is to react emotionally. Misinformation exploits this. Take one breath and decide if this content needs verification before you amplify it.
I. Investigate the Source
Who created this content? What do other trusted sources say about them? A slick website design does not equal credibility. Open a new tab and search for the source's reputation before evaluating their specific claim.
F. Find Better Coverage
Is a trusted news outlet covering the same story? If Reuters, AP, or the BBC are not reporting it, ask why. The absence of coverage from established outlets is itself a signal. If multiple credible sources agree, the claim is likely true.
T. Trace to Original
Find the original source of any claim, quote, or image. Social media posts often recycle content stripped of context. The original may tell a very different story. For videos, BSmeter searches for evidence sources on the weakest claim (the 3 weakest on Basic and Premium) — a useful starting point, but still trace the original yourself.
Let BSmeter Do the First Pass
For videos and podcasts, paste a YouTube, TikTok, Instagram Reel, or podcast link into BSmeter (free account). It transcribes what's said, pulls out the claims, scores each one 0-100, names the framing moves, and finds sources for the weakest one (the 3 weakest on Basic and Premium). Results are AI-generated and can be wrong — then use the techniques above to go further.