Free guide
How not to be misled by media.
The 7 tricks viral videos use to lie to you — and the 60-second manual check for each.
One narrow promise, fully kept: by the end of this page you will know the seven manipulation patterns that carry most viral misinformation, and you'll have a do-it-yourself check for each one. No jargon, no lecture, no “kids these days.” Just the tells, and what to do about them.
The stakes
Why your feed is worth defending.
Here is the situation, with sources, because that's how we do things around here.
One in five American adults now regularly gets news from TikTok — among 18-to-29-year-olds it's 43% (Pew Research Center, 2025). Meanwhile the FTC logged $2.1 billion in reported losses in a single year to scams that started on social media — eight times the 2020 figure — and people aged 18–29 were the most likely to say the scam that got them started there (FTC Data Spotlight, 2026). Not grandparents. Us.
And the underlying physics are against you: a large MIT study of Twitter found false news spread significantly farther and faster than true news — falsehoods were about 70% more likely to be reshared (Vosoughi, Roy & Aral, Science, 2018). Lies are engineered to travel. Truth commutes.
You cannot fix the feed. You can fix how you read it. Seven tricks cover most of what's aimed at you, and each has a check you can run in about a minute. Here they are.
Trick 1 of 7
The rented Lambo
Every “I got rich, and so can you” video leans on visual proof: the car, the penthouse balcony, the airport lounge. Here's the problem — everything in the frame can be rented by the hour. Lamborghinis rent by the day. Mansions rent for photo shoots. Stacks of cash can be a bank withdrawal filmed before redeposit, or props.
Lifestyle footage is production design. It proves a budget for one afternoon, not a business.
The 60-second check
Ask, “is there anything in this video that could NOT be rented, borrowed, or staged?” Audited returns? A named, checkable company you can look up in a state business registry? A verifiable client? Almost always, no. If the Lambo is the only evidence, the Lambo is the confession — it's all they've got.
Trick 2 of 7
Manufactured certainty
Real experts hedge. They say “usually,” “it depends,” “in most cases.” That's not weakness — that's what knowing a subject actually sounds like, because reality has exceptions.
Salesmen don't hedge. “Guaranteed.” “Works every time.” “They don't want you to know this.” Certainty feels trustworthy, which is precisely why it's the con's favorite costume — and part of why false news outruns true news: certainty is engaging, nuance is boring.
The 60-second check
Count the hedges. Zero hedges plus a big promise = flip your default to skeptical. Then search the claim's key phrase plus the word “evidence” or “study” and skim what institutions (not other creators) say. A creator who says “this worked for me, your mileage may vary” has earned more trust than one who promises the universe. Confidence is a delivery style. Evidence is a source you can go read.
Trick 3 of 7
The invisible funnel
One question filters most bad advice on the internet: what happens if you believe this person?
If the answer is “nothing — the information just sits there,” you're probably watching someone share what they know. If the answer is “I click the link in their bio, join the Discord, buy the course, book the call” — you're not watching content. You're watching an ad that doesn't disclose it's an ad. The video is free because the video is the funnel.
This doesn't make every seller a liar. It means the incentive is now part of the claim. “This supplement changed my life [affiliate link]” and “this supplement changed my life” are two different sentences.
The 60-second check
Open the profile. Bio link? Course? “DM me ‘MENTOR’”? Now re-watch the video knowing what it's selling, and ask whether the claim survives the incentive. Advice that ends in a checkout page gets the ad-read standard of trust, not the friend standard.
Trick 4 of 7
The manufactured deadline
“Price goes up at midnight.” “Only 3 spots left.” “This video is getting taken down soon.”
Real deadlines exist — tax day, application windows, an actual event. But urgency in a sales pitch is usually manufactured, and it's manufactured for one reason: urgency is the enemy of checking. Every second you spend verifying is a second you might walk away, so the pitch is built to make verification feel like losing your spot.
The 60-second check
Ask “why the rush?” and then simply wait one hour. Almost nothing legitimate punishes you for taking sixty minutes to verify it. If the deal can't survive an hour of scrutiny, that fact is the verdict. “Why the rush?” is a complete question. Silence is a complete answer.
Trick 5 of 7
The illegal shortcut
Anatomy of the credit-identity pitch. You've seen it: “Bad credit? Banks hate this. Get a CPN and start over with a clean file.”
What's actually being sold: a “credit privacy number” — usually a stolen or fabricated nine-digit number you're told to use instead of your Social Security number on applications. There is no legal parallel credit identity. Using a false number on a credit application is fraud — yours, not the seller's, since you're the one signing. The seller charges you hundreds of dollars to commit it.
The legit free alternative (a real one always exists — that's the tell): free credit reports at annualcreditreport.com, and free direct disputes with the bureaus. Slow and boring. Also legal and real.
The 60-second check
Any pitch whose product is “a shortcut around a system that keeps records” is selling either fraud or fiction — systems that keep records are hard to shortcut; that's what records are for. Search the product's name plus “FTC” or “scam.” If the free legitimate route exists, the paid secret route is the trap.
Trick 6 of 7
The fake giveaway
“First 500 people to comment get $750.” “Tagged in a giveaway? Claim your prize — just cover shipping.”
What the giveaway actually harvests: your comment (free engagement that trains the algorithm to spread the post), your DM reply (personal info for resale or later, more targeted scams), or your “shipping fee” (your card number — the prize was never real, but the charge is). Shopping-style scams were the single most-reported social media scam category in the FTC's latest data (FTC, 2026).
The 60-second check
Run the math out loud. 500 people times $750 is $375,000 — nobody hands $375,000 to strangers for comments. Then look for what real sweepstakes always have: published rules, a named sponsor, and no fee to receive a prize. In the U.S., “pay to claim your prize” is not a red flag — it's the whole crime, described.
Trick 7 of 7
AI slop news
The newest one: video “news” that was never reported by anyone. A synthetic voice over stock or generated footage, announcing an arrest, a death, a new law, a miracle cure. No reporter, no outlet, no date — just vibes and a voice that never breathes wrong.
The 60-second check
In rough order of reliability:
Look for a named source in the video itself. Real news says who reported it. Slop says “reports are emerging” and “sources say” — sources who are never, ever named.
Search the story. Genuinely big news is covered by multiple outlets within hours. If a “huge story” exists only in one faceless video, the story is the video.
Notice the emotional pitch. Slop is optimized for outrage or awe, because those get shares. Notice when a video makes you feel before it tells you anything checkable.
The uncanny details — hands, text on signs, perfectly even vocal pacing — are fading fast as tools improve, which is why the first two checks, which are about sourcing, will outlive them.
The shared toolkit
How to read a source in 30 seconds.
Five of the seven checks end with you looking at a source. “Do your own research” is useless advice if nobody shows you the fast version. Here it is — three questions:
- 1
WHO is publishing this?
An institution with something to lose (a journal, an established outlet, a government agency, a hospital) or a site nobody's heard of wearing a serious-sounding name? Reputations that took decades to build are the collateral behind a claim. Anonymous sites have no collateral.
- 2
WHEN is it from?
A 2016 statistic in a 2026 video isn't automatically wrong — but the creator chose an old number, and choices have reasons. Check the date before you repeat the stat.
- 3
WHAT does it actually say?
This is the big one. Click through and see whether the source claims what the video claims — or something much weaker. “Study finds X may be associated with Y in mice” becomes “X CURES Y” by the time it reaches your feed. The gap between the source and the claim is where most misleading content lives.
Bonus rule at no charge: a screenshot of a headline is not a source. Headlines get cropped, altered, and generated. If you can't click it, it doesn't count.
The fine print, printed large
The honest part.
You now have the complete kit: seven tricks, seven checks, one source-reading routine. It works. Run it on the next video that smells off and you'll catch most of what's aimed at you.
Here's the honest part: done properly — finding the sources, reading them, checking the dates and the gaps — the routine takes ten to twenty minutes per video. That's fine for the one video that really matters. It is not fine for a saved folder with 200 videos in it, and you will not do it at 11pm for a 40-second clip about seed oils. Nobody does. That's not a discipline problem; it's an arithmetic problem.
None of this requires becoming a cynic who trusts nothing, either. Cynicism is just gullibility with the sign flipped — “everything is fake” is as lazy as “everything is true.” The goal is a habit: before you buy, repeat, or believe — find out where the claim came from. Who profits. What the source actually says. Why the rush.
The pitch, now that you can check it
The manual routine works. Nobody runs it 200 times.
We built the machine that runs this exact routine for you: paste any TikTok, Reel or Short into UnscrollMe and about a minute later you get the takeaway plus a verdict on every claim — with the sources linked, so you can run the WHO/WHEN/WHAT test on us the same way this guide just taught you to run it on everyone else. When we can't verify something, we say “needs context” instead of bluffing.
You know how to do this by hand now. This is the version that keeps up with your feed.
Create a free account — 5 checks a month, no cardNo ads, ever. Know what's real in what you save.
Not ready to sign up?
Get this guide in your inbox to keep — and nothing else unless you ask.