Fortinet auth holes, China distills AI, Mythos vulnerability
Summary
Host Sarah Lane covers Fortinet's critical authentication flaws and a US accusation that Chinese AI firms are distilling American models. The episode highlights a human bottleneck in AI vulnerability triage, Anthropic researcher Jacob Coxon's departure over safety concerns, and a Veridigm breach exposing patient data. Additionally, it details Microsoft's new school AI privacy rules and the rapid adoption of the Blue Moon exploit kit by suspected Chinese espionage groups.
Topics discussed
Fortinet patches critical authentication flaws
US accuses Chinese AI firms of model distillation
AI vulnerability findings hit human bottleneck
Anthropic researcher warns on AI safety risks
Sponsor: ThreatLocker AI security demo
Gambling Goblin compromises Brazilian gov servers
Veridigm breach exposes patient data
Microsoft signs school AI privacy rules
Blue Moon exploit kit targets patch gap
Outro and show resources
Listen ad-free on Castria