Hacking Humans

Lost iPhone, found trouble.

Nov 20, 2025
This week, the hosts dive into alarming social engineering scams, including China's crackdown on a violent fraud gang and the extradition of a key scam figure. They discuss an unsettling sextortion method involving AI-generated images from unsolicited FaceTime calls. A debate erupts around the legitimacy of an AI-driven cyber-espionage claim by Anthropic amid skepticism from researchers. Plus, there's a warning about phishing attacks using lost iPhone contact info and mobile shopping threats during the holiday season.
Ask episode
AI Snips
Chapters
Transcript
Episode notes
INSIGHT

Public Data Amplifies AI Extortion

  • Attackers combined an opportunistic FaceTime photo with AI image editing and publicly scraped contact lists to make extortion credible.
  • Publicly available people-search data makes targeted social blackmail trivial and believable.
ADVICE

Avoid Showing Camera To Unknown Callers

  • Answer FaceTime calls without enabling your camera when possible to avoid exposing live images to unknown callers.
  • Be cautious about accepting unsolicited video calls, especially in semi-public settings like your car.
INSIGHT

AI Used To Scale, Not Reinvent, Attacks

  • Anthropic reported attackers used its Claude model to orchestrate multi-step intrusions by breaking tasks into small prompts and chaining actions.
  • Researchers caution this is more automation of existing tooling than a fundamentally new autonomous offensive AI capability.
Get the Snipd Podcast app to discover more snips from this episode
Get the app