AISafetyWire (@AISafetyWire) profile picture

Real AI scams, explained like a story. Spot them before they spot you. Plain language. New stories at https://nitter.cf/t.co/Z0LQjg8xM2

Bengaluru, India
Joined November 2026
AI Safety Wire: the AI news that affects real people, told as stories. Scams, deepfakes, fake books and hackers. Plain language. One protective habit in every story. Follow so the next one finds you. (1/2)
1
17
Reuters tests allegedly found Meta’s AI agent Muse could reach Signal messages on an Android phone and a Mac, using permissions users granted at setup. The same requests were blocked on an iPhone. A thread on what that means 🧵
1
4
Why Signal’s encryption didn’t stop it: Encryption protects a message while it travels. It can’t protect it once it appears on your screen. Anything you’ve allowed to watch your notifications or screen can read it there.
1
2
The Hacker Who Doesn't Sleep FBI WARNING: AI HACKING TOOL ARTEX: released July, taken down 8 October. A former pentester allegedly used it on 90+ organisations and 8.2 million records. Arrested 15 October. The scary part is not the hack. It is how little skill it needed. (1/2)
1
1
98
The Library Full of Books Nobody Wrote FAKE BOOKS, REAL MONEY Allegedly non-existent authors. ISBNs borrowed from real books. Reportedly 70,674 suspect titles and about Rs 7 crore in Tamil Nadu. An ISBN is the number that identifies a book. Here is how this one was gamed: (1/2)
1
8
DEEPFAKE INVESTMENT SCAM A video of the Finance Minister "recommending" an app. Rs 49.2 lakh gone in about ten weeks for one woman. Rs 1.41 crore over 30 transfers for another man. Would you check before sending the first Rs 1,000? (1/2)
1
3