How-To & Security
A calm, practical guide to thinking critically online: checking sources, using reverse image search, recognizing the signs of AI-generated or manipulated media, and slowing down the urge to share.
Something lands in your feed and your stomach reacts before your brain does. A shocking headline. A clip of a public figure saying something outrageous. A photo that seems to prove a point you already believed. Your thumb hovers over the share button, because surely people need to see this. That moment — that small gap between feeling and sharing — is where almost all the damage of fake news and deepfakes either happens or gets stopped.
I want to be reassuring here, because this topic can tip into hopelessness fast. You don't need special software or a journalism degree to navigate it. You mostly need a few calm habits and a willingness to slow down by about thirty seconds. Let's build those habits together.
Before any technical trick, there's a mental one, and it does more than all the others combined. Notice your emotions. Misinformation and manipulated media are engineered to bypass your thinking and go straight for your feelings — outrage, fear, vindication, delight. That's not a side effect; it's the whole design. Strong emotion makes you share fast, and sharing fast is exactly what spreads it.
So flip the cue. When something makes you feel a powerful jolt and an urge to share right now, let that urge be your signal to stop instead. The stronger the reaction, the more it deserves a pause. Ask yourself a plain question: do I actually know this is true, or does it just feel like it should be?
The strength of your urge to share something immediately is a pretty reliable measure of how much you should slow down and check it first.
That pause costs you almost nothing and protects you from being the person who accidentally passes along something false to everyone you know. It's the cheapest, most effective tool you have.
Once you've paused, do a little tracing. A huge amount of online misinformation travels as screenshots, quotes ripped from context, and "my friend's cousin works there" summaries. None of those are the actual source. They're a copy of a copy, and copies are easy to alter or invent.
So go looking for the original. A few questions guide you well:
If a claim is real and significant, you can usually find it reported in more than one credible place. When something only exists as a single dramatic post with no traceable origin, that absence is itself a warning. And be a little wary of headlines that don't match the article beneath them — sometimes the body text quietly walks back the alarming claim in the title, counting on you not to read that far.
Photos feel like proof, which is exactly why they're so often misused. A genuine image from one event gets relabeled as another. A picture from years ago resurfaces as if it happened yesterday. The image itself may be completely real while the story attached to it is entirely false.
This is where a reverse image search earns its keep, and it's easier than it sounds. Most search engines let you search using a picture instead of words — you upload the image or paste its link, and the engine shows you where else that image has appeared online. Suddenly you can see that the "breaking" disaster photo actually ran in a news story three years ago, in a different country, about a different event.
It takes under a minute and it's remarkably clarifying. When a picture is doing a lot of emotional or persuasive work in a post, a quick reverse search is one of the most useful checks you can run.
Now to the part that worries people most: deepfakes and AI-generated video, audio, and images. The honest framing is important here. The technology is improving quickly, so any checklist of "tells" is a snapshot that will age. Treat the signs below as hints that something might be synthetic, never as proof one way or the other.
With images, look closely at the small, fiddly details that generators have historically struggled with — hands and fingers, teeth, jewelry, text on signs, the way reflections and shadows fall. Backgrounds can blur into nonsense or contain warped, melted-looking objects. With video and audio, watch for mouth movements that don't quite match the words, oddly smooth or flickering skin, unnatural blinking, and a voice whose rhythm or emotion feels subtly off.
But here's the more durable advice, the part that won't go stale: lean on context over forensics. Instead of squinting at pixels, ask whether a credible source is also reporting that this person said or did this thing. A genuinely newsworthy video of a public figure won't exist in a vacuum — it'll be covered, confirmed, and discussed by outlets you can trace. A shocking clip that appears only on anonymous accounts, with no reputable source touching it, deserves deep skepticism no matter how convincing it looks. As the fakes get better at fooling your eyes, this kind of sourcing check becomes your steadiest ground.
All of this comes back to where we started: the share button. You don't have to verify everything you ever see — that would be exhausting and unnecessary. The habit to build is narrower and kinder to yourself: before you actively spread something, especially something emotional or consequential, take that brief beat to check.
A simple internal rule works well. If I can't point to where this came from, and I haven't done a quick check, I don't pass it on. You're not being paranoid or cynical by doing this. You're being a considerate node in everyone's information network — the friend who doesn't fill the group chat with things that fall apart on a second look.
One last note, offered gently. The tools and signs in this article will keep shifting as the technology does, so it's worth refreshing your sense of current best practices now and then rather than assuming today's tells will hold forever. And if you ever find yourself targeted by something serious — a convincing impersonation, a manipulated image used to harm or defraud you — that's a moment to involve the right professionals or official bodies rather than handling it alone.
Thinking critically online isn't about distrusting everything until you're cynical and tired. It's about a few calm habits: noticing your emotions, tracing claims to their source, searching an image when it matters, and leaning on credible reporting over your own pixel-squinting. Most of all, it's that small, powerful pause before you share. Build that one habit, and you've already done more than most.
Mara is a digital-security writer who believes good security is a set of small habits, not constant paranoia. She turns intimidating topics — passwords, backups, phishing, privacy — into calm, doable steps. She's helped enough people recover from avoidable disasters to take backups very, very seriously.
Keep reading
A calm, practical guide to online privacy for normal people — what data you quietly leak, the settings worth tightening, and why sharing less is the simplest protection you have.
Mara Lindqvist
Forget the symbols-and-numbers gymnastics. A calm guide to passwords that are actually strong — why length beats complexity, how to build memorable passphrases, and why a manager does the heavy lifting.
Mara Lindqvist