Everyone Needs an AI Radar Now. Train Yours in Minutes a Day
Within one scroll of any feed you now pass AI-generated photos, machine-written captions, synthetic product reviews and the occasional deepfake ad. Most of it is harmless. Some of it is trying to sell you something that doesn't exist, or make you believe something that didn't happen.
The ability to sense what's synthetic — call it an AI radar — has quietly become a baseline life skill, like judging whether a link is safe. The good news: it's trainable, fast, and the training is genuinely fun.
Why You Can't Outsource This to Software
The obvious objection: "won't detector tools handle it?" No, for three reasons.
- Detectors lag generators permanently. Every detector is trained on yesterday's fakes; every generator ships tomorrow's. The gap never closes in the detector's favour.
- They fail unfairly. Text detectors notoriously flag non-native English speakers as machines; image detectors miss anything lightly edited. False accusations do real damage.
- They're never there at the moment of decision. Nobody runs a detector on a WhatsApp forward before believing it. Your radar is the only tool that's always installed.
Software is a second opinion. The first opinion — the one that decides what you share, buy and believe — is yours.
What a Calibrated Radar Feels Like
People with a trained eye don't stare longer; they check differently. A glance at text in the image, edges, lighting (the full list: 9 tells for AI images). A feel for over-smooth, over-structured writing (the text tells). And above all, the reflex that survives even perfect fakes: check the source before believing the content.
Crucially, a good radar is calibrated in both directions. Untrained people make two errors: believing fakes, and — increasingly common — confidently "detecting" AI in things humans made. Both are failures. Training fixes both.
The Science Bit: Feedback Makes the Skill
You've seen thousands of AI images already; they didn't train you, because nobody told you which were AI. Exposure without feedback teaches nothing. Exposure with immediate feedback — guess, reveal, adjust — is how perceptual skills are actually built, from radiology residents to birdwatchers. The loop matters, not the volume.
Two Minutes a Day, Measurably Better
That loop is exactly what AI or Not gives you: a free daily game with 8 rounds a day — real photo or AI image? human writing or machine? — with instant reveals and a streak to keep you coming back. It takes about two minutes.
The typical experience: day one humbles you (most people score far worse than they predicted), and within a couple of weeks the tells start jumping out before you've consciously looked. That transfer is the point — the game trains the glance you'll use on every feed, forward and "urgent" message from now on. It pairs naturally with scam-reflex training, the other half of modern self-defence: You Can Train Your Scam Radar.
Play today's 8 rounds — and check your streak against your confidence.
Cocoon builds free AI tools and runs practical AI training for professionals and teams across Sri Lanka and Southeast Asia. Try the free tool from this article or talk to us about training.