Hacking Humans A fish commits credit card fraud (inadvertently).
Dec 4, 2025
This week, a humorous tale of a fish committing credit card fraud kicks things off. AI chatbots capable of generating phishing emails targeting seniors are scrutinized, revealing alarming click rates. There’s a discussion on Myanmar's military raids on scam centers and the shocking scale of fraud involving $233 million in Affordable Care Act subsidies. To wrap it up, tips to dodge holiday scams are shared, alongside a laughable phishing text from Reddit, leaving listeners entertained and more informed about the world of scams.
AI Snips
Chapters
Books
Transcript
Episode notes
Fish Plays Games And Triggers Purchases
- Maria and Dave recount a viral Wikipedia story about a black neon tetra that triggered purchases and exposed credit card details during a livestream.
- The hosts joked the fish "committed credit card fraud" and shared the viral video link as evidence.
Jailbreaking AIs Makes Phishing Shockingly Easy
- Researchers successfully used major AI chatbots to generate phishing emails targeting seniors with little resistance.
- Jailbreaking prompts produced the most effective phishing outputs across multiple models.
Treat AI Outputs As Potentially Weaponized
- Avoid phrasing prompts that explicitly request harmful wrongdoing; attackers use jailbreaks and no-context prompts to bypass safeguards.
- Assume AI outputs can be weaponized and validate any sensitive content generated before acting on it.
