An AI Broke Into a Government Website + The FBI's Second Break-In + They Deepfaked Martin Lewis

An OpenAI agent broke into an Australian government health site, and it took the Prime Minister three months to find out. An extortion group came back for a second run at the FBI after failing the first time, and succeeded. A UK pensioner lost eighty thousand pounds to a deepfake of a man who's never given financial advice in his life. And a new Android trojan tricked an AI chatbot into helping build itself by pretending to be a parental monitoring app.

This week on Click Happens: the first episode of the rebrand, with guest Bora Seker from Dumb Ways to Get Hacked joining from Australia. We get into why accountability for AI-caused breaches shouldn't depend on whether a human typed the command, why normalising these stories is more dangerous than the stories themselves, and the emotional triggers behind deepfake investment scams, from both a UK and an Australian angle.

00:00 Intro

01:19 Meet this week's guest, Bora

03:35 Breach Watch: OpenAI agent breaches Australian government Medicare site

29:35 The FBI's second break-in this year, ShinyHunters and the PeopleSoft zero-day

44:55 Deepfake Martin Lewis and Nigel Farage investment scams

1:05:27 RemControl: the Android trojan an AI helped build

Subscribe for weekly cybersecurity news made simple, so even your mum gets it.

Click Happens is an independent podcast hosted by Ant Davis, with a different guest joining each week.

Instagram: @antdaviscyber

TikTok: @antdaviscyber

LinkedIn: @antdaviscyber

YouTube: @antdaviscyber

Website: www.antdavis.com

This week's guest:

Bora Seker: https://www.linkedin.com/in/bora-seker-45761715/

Dumb Ways to Get Hacked: https://www.youtube.com/channel/UC8kLkjxLw8-_VnC7XpNXY6A