AI Fact Checker — Is This Actually True?

Paste any claim, headline, or post. Get a truthfulness score from 1–10 with cited sources and a debate trail you can verify yourself.

  1. Paste the claim — A headline, a social post, a friend’s claim. Anything specific you can verify against evidence.
  2. Lenz researches it — Multiple independent sources searched in parallel, evidence collected pro and con.
  3. AI models debate — Two AI advocates argue opposing sides. Then three independent expert models score the evidence.
  4. See the verdict — A 1–10 truthfulness score, a conclusion label, and every cited source. Verify the verification.

The whole process typically takes about a minute or two. Read more about how Lenz verifies claims.

News headlines

Did the senator really say that? Did the study really show that? Verify the claim against the underlying evidence.

Political claims →

Health & science claims

Cancer cures, miracle supplements, viral health advice. Lenz checks the evidence against peer-reviewed research and authoritative sources.

Health claims →

Things you read on social media

Forwarded WhatsApp messages, viral TikToks, Twitter posts that sound too good (or too bad) to be true. Verify before you share.

Check before sharing →

Browse the full library of verified claims.

What does the truthfulness score mean?
Lenz scores every claim from 1 to 10 based on how strongly the available evidence supports it. 9–10 is True, 7–8 is Mostly True, 4–6 is Misleading or partially supported, 1–3 is False. Each score comes with a written conclusion and the sources it relied on, so you can verify the verification.
Is the AI fact checker free to use?
Yes. You get 10 free verifications per month, no signup required to try. Plus and Pro plans unlock more if you need them.
How is this different from ChatGPT or Google?
ChatGPT gives you a confident answer with no sources. Google gives you 5 tabs and an opinion. Lenz researches the claim across independent sources, debates the evidence with multiple AI models, and shows you the full reasoning trail and citations behind the verdict.
How accurate is the verdict?
Accurate within the limits of what state-of-the-art AI and the available sources can determine. Lenz shows its work, names the sources, and is honest about uncertainty. A score of 6 with “Mostly True — but the study was small” is more useful than a confident binary verdict. Read about why Lenz exists and the editorial choices behind the process.
What kinds of claims work best?
Specific factual claims with verifiable evidence: news headlines, statistics, health and science claims, historical statements, viral social posts. Subjective opinions and predictions about the future don’t have evidence to verify against.
Can I share the result?
Yes. Every verification has a public page with the verdict, score, and sources. Share the link in a group chat, on social media, or with anyone who asked the same question.

Stop guessing. Start verifying.

Paste any claim, headline, or post. Get a sourced verdict you can trust and share.

Verify a claim now