Zuckbot, Rockstar, Klaude, Browsers Galore, Microsoft 365, ATC, Kieran Human and more - Kieran Human - SWN #572

The Security Weekly News episode 572 dives into a cascade of high-stakes cyber threats, from the massive data wipe at Striker Medical—where attackers allegedly exploited a compromised Microsoft 365 account to erase tens of thousands of machines—to a fake Claude AI site distributing the PlugX malware via a deceptive MSI installer. The episode underscores how attackers now operate in 3D: compromising third-party vendors like Anodot to breach high-profile targets like Rockstar Games, exposing 78.6 million records including in-game economies and support tickets. A stark warning emerges: even with MFA, credentials can be bypassed through email forwarding, SIM swapping, or malicious browser extensions—especially AI-powered ones, which are 60% more likely to contain known CVEs. The show stresses that zero trust isn’t optional; it’s essential. Kieran Heumann from ThreatLocker emphasizes that assuming breach is the foundation of resilience, and that tools like ZTCA (Zero Trust Cloud Access) can block unauthorized logins even from privileged accounts by enforcing device trust. The episode ends with a satirical yet chilling look at Meta’s AI Zuckbot—training an AI avatar of Zuckerberg to replace human leadership—hinting at a future where AI billionaires outlive their creators, and the real danger isn’t just data theft, but the erosion of human oversight in critical systems.