Open an AI search tool that shows citations (Perplexity, Microsoft Copilot, or Google's AI Overview), ask whether the specific claim is supported by reliable sources, then read at least one of the cited sources directly. The key is checking what the AI links to, not just accepting its summary.
A suspicious headline lands in your feed. Your first instinct might be to share it — or argue about it — before you've verified anything. This workflow gives you a way to check in about 60 seconds, using AI tools you likely already have access to.
You don't need an account, an app, or any technical skills. You just need a browser.
Isolate the exact claim
Before you fact-check anything, be clear about what the claim actually is. Headlines are often vague or misleading even when the article underneath is more nuanced.
Ask yourself: What specific fact would have to be true for this headline to be accurate? Write it down or copy it to your clipboard. A headline like "Scientists Say Coffee Causes Cancer" would need a specific study from a named institution to be real. That's what you'll search for — not the headline itself.
Open an AI search tool that shows citations
Standard chatbots like a basic ChatGPT window aren't ideal for fact-checking because they don't always pull from current sources. Instead, open one of these:
- Perplexity AI — free at perplexity.ai, shows inline citations by default
- Microsoft Copilot — free at copilot.microsoft.com, connected to the live web
- Google AI Overview — appears automatically at the top of many Google searches
All three pull from live web sources and show you where the information came from. That's what separates them from a regular chatbot for this purpose.
Ask a specific verification question
Don't just paste the headline. Ask a question that forces the AI to find sources. A reliable prompt looks like this:
Is it true that [specific claim]? Show me which news outlets or official sources confirm or deny this.
For example:
Is it true that the FDA banned food dye Red 3 from all foods in 2025?
Show me which official sources or major news outlets reported this.
The more specific your question, the more useful the answer. Avoid yes/no questions — you want sources, not just a verdict.
Read the citations, not just the AI summary
This step is the most important one. The AI summary can still be wrong or incomplete. What matters are the sources it links to.
Look for citations from:
- Major newspapers (New York Times, Washington Post, Guardian, AP, Reuters)
- Government websites (.gov domains)
- Academic institutions (.edu domains)
- Established broadcast networks (BBC, NPR, PBS)
If the sources are all from sites you've never heard of, or if there are no citations at all, treat the claim as unverified — regardless of how confident the AI summary sounds.
Cross-check with one trusted outlet directly
Open a new browser tab and search the claim on a site you already trust. A reliable shortlist:
| Outlet | Best for |
|---|---|
| AP News (apnews.com) | Breaking news, US and international |
| Reuters (reuters.com) | Business, global news |
| BBC News (bbc.com/news) | International stories |
| NPR (npr.org) | US politics, health, science |
| Snopes (snopes.com) | Viral claims and urban legends |
| PolitiFact (politifact.com) | Political statements |
| FactCheck.org | US election and policy claims |
If none of these have covered a story that supposedly happened recently, that's a strong signal to pause before sharing.
Decide and act
After your check, you have three options:
It checks out — share if you want; you could even mention you verified it.
It doesn't check out — don't share. If a friend already shared it, a private message explaining what you found is usually well-received.
You're not sure — wait. A real story will still be real tomorrow. There's no cost to waiting; there's a real cost to spreading false information.
If you found the story on social media, most platforms let you report it under a "false information" or "misleading content" option.
What to Try Next
Now that you have a fast verification workflow, AI-Generated Fake News: Spotting Content Farms explains the broader patterns to watch for — so you'll start recognizing suspicious sources before you even click. If you're also wondering whether text you're reading was AI-written rather than human-written, How to Tell If Text Is AI-Generated covers the most reliable signals.
