**AI Granny Destroys Phone Scammers By Wasting Their Time, And Honestly, We Stan**
Look, I know we’ve all been doomscrolling through the AI news feed and it’s usually the same depressing cycle: “AI took my job,” “AI wrote a creepy poem about my soul,” “AI told a lawyer to cite cases that don't exist.” It’s exhausting. But for once, the robots are actually using their terrifying powers for good, and it’s because some genius at British telecom giant O2 decided to weaponize a chatbot against the absolute worst scum of the earth: phone scammers.
Enter “Daisy.” Not a real person. Daisy is an AI-powered grandmother. She is the hero we didn't know we needed in this godforsaken timeline. O2 trained her to be the ultimate time-wasting weapon. She talks slow. She gets confused. She asks you to repeat yourself. She mentions her cat, Fluffy, approximately 47 times per conversation. She is, in essence, a digital honey trap designed to keep scammers on the line until their own phone batteries die of old age.
And it’s working.
The reports coming out of the UK are hilarious. Scammers are calling “Daisy” thinking they’ve hooked a live one—a sweet, technologically illiterate old lady who will definitely hand over her credit card info for a “free” antivirus software subscription. Instead, they get a woman who spends 20 minutes trying to find her glasses so she can read the screen, then another 15 minutes talking about how “the internet box” keeps making a buzzing noise. By the time the scammer realizes they’ve been played, they’ve wasted half an hour of their shift for exactly zero dollars. That’s a net loss for the scammers, and a massive W for the rest of humanity.
This is the kind of petty, beautiful, chaotic energy we need more of in AI development. Forget using machine learning to cure cancer for a second (we’ll get there, nerds). Let’s use it to ruin the day of the guy who keeps calling me about my car’s extended warranty. O2 released a statement that basically said, “Yeah, we deployed this bot, she talks to about 200 scammers a week, and she keeps them on the line for an average of 40 minutes.”
Forty minutes. Do you understand how long that is in scammer time? That’s like a dog year. That’s three full episodes of *The Office* (US version, obviously, the superior version). That’s enough time for the scammer’s boss to yell at them for not hitting their quota. The psychology is brilliant. These scammers operate on volume. They need to get 100 people on the line, scare 50 of them, and scam 10 of them to make rent in their Mumbai call center (or, you know, their basement in Florida). By wasting 40 minutes on “Daisy,” they miss the window to scam an actual human being.
But let’s get into the real meat of why this is going viral. It’s not just the technology. It’s the *personality*.
Developers didn’t just make a robot that says “hello.” They gave Daisy a backstory. She loves knitting. She has a cat. She says things like, “Oh, dear, is this the tax man? I think I put the check in the mail.” She’s programmed to take the bait, but slowly. It’s like watching a nature documentary where the predator thinks they’ve cornered the slowest gazelle on the savannah, but the gazelle is actually a master ninja wearing a furry costume.
Social media is eating this up. Reddit threads are popping up with people begging O2 to open-source the bot. Twitter (sorry, X, I’m not calling it that) is flooded with fan art of Daisy as a Terminator-style character, but instead of a shotgun, she’s holding a landline phone and a ball of yarn. Someone already made a deepfake of Joe Biden using this bot to prank call Trump. It’s the most united Americans have been since 9/11.
Of course, the cynics are out. “What about privacy?” they whine. “What about the energy cost of running a chatbot?” My brother in Christ, the scammers are stealing life savings from retirees. I think we can spare a few server racks to run a digital granny with a bad hip. The scammers are the ones violating privacy. Daisy is simply *consensually* wasting their time. It’s the most ethical use of AI since someone taught it to write a breakup text for your emotionally unavailable situationship.
There’s also the hilarious irony that the scammers themselves are getting a taste of their own medicine. They rely on the human element—confusion, fear, politeness. Daisy weaponizes that exact same confusion. She’s basically the “I am something of a scientist myself” meme, but instead of creating a monster, she’s creating a nuisance. And honestly, a nuisance is exactly what these people deserve.
The real kicker? This is just the beginning. Imagine a future where every phone number is protected by a personal AI granny. You forward your calls to her. She screens them. If it’s your mom, she patches you through. If it’s “Microsoft Support” calling about a virus on your Mac (which doesn’t even make sense, but okay), she keeps them talking about the weather in Liverpool for an hour. Then she hangs up and says, “Bye bye, love.” It’s the ultimate firewall.
Some people are saying this is a slippery slope. “What if the AI learns to scam *us*?” Look, we’re already being scammed by AI-generated voices of our loved ones. That’s terrifying. But for once, the shoe is on the other foot. The scammers are the marks. Let them feel the burn.
This is the kind of AI news that restores a tiny, cynical sliver of faith in humanity (or at least in British telecoms). It’s not about world domination. It’s
Final Thoughts
The relentless churn of AI news reveals a double-edged truth: we are rapidly perfecting tools that can mimic human cognition, but we remain profoundly naive about their societal ripples. While each breakthrough promises unprecedented efficiency, the real story lies not in the algorithms themselves, but in the widening gap between technological velocity and our glacial pace of ethical regulation. Ultimately, the coming year will test whether we can steer this juggernaut toward genuine human benefit, or simply watch it deepen the divides it was meant to bridge.