The call that sounded like Nan's grandson

Margaret is 68 and lives in Geelong. On a Thursday afternoon her landline rings, and the voice on the other end is unmistakably her grandson Ethan: the same accent, the same slightly rushed way of talking, even the little laugh he does when he is nervous. He sounds shaken. Nan, I have crashed the car. Nobody is hurt but it is bad, and if I do not pay the other driver $3,000 today he is going to the police and I will lose my licence. Please do not tell Mum.

The voice explains that his phone was damaged in the crash, so he is calling from a borrowed one. Minutes later a text arrives from an unknown number: It is me Ethan, this is the number to use for now. The easiest way to pay is gift cards, the other driver's mechanic accepts them. Get $3,000 in gift cards from the supermarket and read me the codes. Margaret's heart is pounding. She has her keys in her hand and is halfway to the door.

Here is how the scam was built. Three weeks earlier, Ethan posted a 20 second video to his public social media: him laughing with mates after football, talking to the camera for most of it. That clip was all the scammers needed. AI voice cloning tools can copy a voice from a few seconds of clear audio, and Ethan's account also told them his name, his town, his rough age and that he drives. Margaret's name and number came from ordinary public records and older data leaks. Nothing about this required hacking anyone. It was assembled entirely from what was already public.

Notice the pressure techniques, because they are the same in every version of this scam. Urgency: pay today or the police get involved. Secrecy: do not tell Mum, which removes the one check most likely to break the scam. Emotion: a loved one in trouble, which crowds out careful thinking. An explanation for every hole: the damaged phone explains the unknown number, the mechanic explains the bizarre payment method. And the payment itself, gift card codes, was chosen because reading codes over the phone is like handing over cash. It cannot be reversed.

It is worth understanding why gift cards in particular. Had Margaret read those codes out, the value would have moved the moment the words were spoken. Within minutes the codes would have been redeemed or resold, most likely overseas, leaving no transaction to reverse, no account to freeze and no bank able to claw anything back. Scammers choose payment methods the way burglars choose windows, and every one of their favourites, gift card codes, cryptocurrency, instant transfers to strangers, shares the same property: once the money moves, it is gone.

Standing at the front door, Margaret hesitates. The gift cards bother her. Ethan is sensible about money, and something about paying a mechanic in supermarket gift cards does not sit right. So she does the one thing the script tried to prevent: she rings Ethan's actual number, the one saved in her phone for years. He answers on the third ring, out of breath, from football training. There was no crash. There is no other driver. The voice on the landline was a machine.

The spell breaks the instant a second channel opens. That is the lesson inside the lesson: the clone was perfect, the story was airtight, and none of it survived one phone call to a number the scammers did not control. Margaret did not out think the scam. She just verified through a channel she already owned, and that was enough.

Understand also that Margaret was not chosen because she was special. Scams like this run at industrial scale, with software working through lists of numbers and voice clones generated to order, so the cost of each call is close to nothing. The scammers do not need most calls to work. If one call in hundreds gets through, the operation is profitable, which is why the calls keep coming and why the defence cannot be hoping you are never dialled. The defence is being the call that fails, and Margaret shows how, with one habit that works every time.

That evening, Ethan helps Margaret report the call at scamwatch.gov.au, including the unknown number from the text. Scamwatch uses reports like hers to warn the public and track new scam patterns, so the report matters even though she lost nothing. Ethan also sets his social media to private and trims old public videos. And the whole family agrees on a safe word, so that any future distressed call, however real it sounds, gets one quiet question the machines cannot answer.