AI-powered text spam is getting slicker, cheaper, and harder to spot

AI-powered text spam is getting slicker, cheaper, and harder to spot

AI has helped drive a new wave of spam and scam texts, with AI-powered scam texts surging 1,210% in 2025, far outpacing the 195% growth in traditional fraud. Projected losses reach $40 billion by 2027. One especially common pattern is the "wrong number" scam, where AI helps scammers zero in on area codes and build credible narratives that trap victims. They also lean on fewer spelling and grammar mistakes, context-specific details from breached data, and phone number spoofing so the text looks legitimate. The targeting often starts with data brokers, dark web purchases, and data breaches, and once that information is out there it is extremely hard to remove. AI then helps scammers generate thousands of unique text variations in seconds, so they can blast messages at scale and slip past filters that catch identical spam.

Gobble's Take: If a text feels oddly specific but slightly off, treat it like a costume, not a confession. Source: Perplexity Search (community news)

In Case You Missed It

Yesterday's top stories:

• Fake invoices for planning matters in Los Angeles
• AI deepfake voice clones are fooling families and banks
• Government imposter scams are evolving into in-person pickup demands