China hits fast-forward on AI security, Hacking Cat shows its claws, Vite servers spill cloud secrets
Summary
This episode covers China’s pushback against US calls to slow AI development and its new safety framework. It details Hacking Cat’s shift to destructive ransomware attacks on Russian targets and a mass scanning campaign stealing cloud secrets from exposed Vite servers. The show also reports on a malicious Twitch extension leaking OAuth tokens, enterprise concerns over AI data retention, and a critical GitLab path traversal flaw. Finally, it highlights a rapid Marimo notebook exploit and Revolut’s breach caused by a sophisticated government impersonation scam.
Topics discussed
China pushes back on US calls to slow AI development
Hacking Cat shifts to destructive attacks on Russian targets
Mass scanning campaign targets V-Server cloud secrets
Twitch extension leaks OAuth tokens to Russian proxies
Sponsor segment: Vanta
Big Tech restricts employee use of advanced AI models
GitLab critical path traversal flaw exploited
Human attacker breaches Marimo notebook in 8 seconds
Revolut falls for government impersonation scam
Outro and feedback
Listen ad-free on Castria