CyberWire Daily
CyberWire Daily

Nov 19, 2026 · 36 min

Listen from 2:17

Listen at 2:17

Actively exploited flaws collide with exposed AI infrastructure

Four flaws, no time to stall.

The episode links urgent vulnerability response with a newer risk: AI systems and agents expanding the attack surface before organizations can govern them.

3 key takeaways
  1. 1CISA adds four actively exploited flaws as attackers probe Java for Shell and vendors rush patches across critical infrastructure.
  2. 2Rapid7 finds thousands of unauthenticated AI and notebook servers exposing sensitive artifacts and enabling unauthorized cloud-resource consumption.
  3. 3SpecterOps’ Sage shows AI can assist offensive security, but unreliable outputs and autonomous actions demand human oversight.

Don't miss

Russell Van Tyle explains how Sage connects AI assistance with Mythic and BloodHound data while retaining human approval over risky actions.

The brief

CISA adds four actively exploited vulnerabilities to its catalog while threat actors scan for Java for Shell, compressing the time defenders have to respond.

Ransomware arrests in Ukraine, Edge Lockdown Mode, and urgent F5, Drupal, Splunk, and PHP fixes show how quickly exposed infrastructure becomes operational risk.

Rapid7 finds thousands of AI and notebook servers without authentication or HTTPS, with some exposing sensitive artifacts or allowing attackers to consume cloud resources.

SpecterOps’ Russell Van Tyle discusses Sage, an experimental system that uses AI, Mythic, and BloodHound data to help operators identify and pursue attack paths.

The interview’s central constraint is reliability: Sage can work in supervised or autonomous modes, but confident errors, loops, and dual-use capabilities keep humans responsible.

A final discussion rejects an imminent superintelligence explosion while stressing that poorly constrained agents and uneven AI systems already create serious security risks.

Weekly email

State of the Pod, every Monday

The week's most-discussed names and stories from podcasts, with the moments behind them. Free, five minutes, unsubscribe in one click.

Listen to the full episode and explore every guest, topic, and moment on PodLume.

Actively exploited flaws collide with exposed AI infrastructure · PodLume