27/10/2026
Real or AI? These days you can’t always tell at a glance, and that’s exactly what attackers are counting on.
AI has made deception cheap and convincing: a cloned voice, a familiar face on video, a flawless email that knows just enough about you to seem legitimate. The old giveaways are disappearing clumsy grammar, odd phrasing, dodgy logos.
What hasn’t changed is that a short pause still beats a fast scam. Take a second to tell the difference.
Five habits worth building this month:
- Don’t trust a voice or face alone. Audio and video can be cloned convincingly now, so treat an urgent request as unverified until you’ve checked it another way.
- Check your privacy settings. The defaults weren’t chosen with you in mind, so take a second to decide who can see you.
- Call back on a number you trust. If a call or message pushes you to act fast, hang up and ring back on a number you already have, never the one they give you.
- Check what personal data you share with AI chatbots. Anything you paste in can be stored or exposed, so keep client details, credentials and sensitive data out of it.
- Take a second to check your AI answer. AI sounds just as confident when it’s wrong, so verify anything you’d act on before you rely on it.
Taking a second also means not taking your own cyber security on faith. The surest way to miss a gap is to let the people who built your systems be the only ones who check them, a bit like marking their own homework.
A second is a small price for staying secure.