
Hacking Humans The evolving face of AI deepfakes.
Mar 26, 2026
They talk about AI-generated deepfakes and the rise of recruiting real faces for large-scale scam operations. They cover automated phishing tools that craft realistic lures for training. INTERPOL’s assessment of evolving AI-enabled financial fraud gets attention. The conversation also explores the growing refund fraud economy and a humorous scambait thread from Reddit.
AI Snips
Chapters
Transcript
Episode notes
Backyard Chicken Coop Troubles
- Joe Kerrigan described building a new chicken run and dealing with misbehaving hens like Dottie scattering feed during “bath” antics.
- He added perches and an old run to reduce bullying of a smaller hen named Speckleface and now collects ~4 eggs a day.
LinkedIn Speak Can Be Automagically Polished
- Kagi's LinkedIn translator turns blunt English into over-polished corporate speak, revealing how generative AI can produce inauthentic social posts.
- Joe tested it with a complaint about a manager and got an absurdly glowing LinkedIn-style rewrite.
Use AI Phishing Demos For Training
- Use the Civ AI phishing-demo tool to teach people to spot phishing by generating convincing, persona-targeted scam emails.
- The tool pulls Wikipedia bios and creates tailored phishing lures and landing pages, including fictional characters for safe practice.
