
Nov 16, 2026 · 13 min
Gateway attacks and poisoned AI tools expose fragile defenses
PHP2Shell attacks on CheckPoint gateways, Stolen AI key behind Cursor attack, Microsoft patches break printing and activation
The episode connects destructive infrastructure attacks, compromised developer tooling, vulnerable industrial systems, and unreliable security patches into one widening risk picture.
- 1Attackers used PHP flaws and a Check Point gateway zero-day to plant web shells and wipe VMware ESXi servers.
- 2A stolen Fireworks AI key exposed coding-agent data and enabled trojanized Cursor extensions that reached thousands of users.
- 3Microsoft’s November patches caused operational failures while fixing more than 1,259 vulnerabilities, including an exploited Outlook flaw.
Don't miss
The Fireworks compromise shows how one stolen AI service key could expose internal traces and turn a coding marketplace into a malware channel.
The brief
Attackers exploited PHP vulnerabilities and a Check Point gateway zero-day, moving from web shells into VMware ESXi environments and deploying the Destroy_v6 wiper.
A stolen Fireworks AI access key exposed coding-agent traces and credentials, then helped attackers publish trojanized Cursor extensions downloaded by as many as 5,000 users.
The episode turns to supermarket refrigeration controllers, where 14 root-access flaws left critical systems exposed and some devices were already being probed or manipulated.
A proposed AI breach safe harbor could encourage faster reporting, but only if it excludes delayed disclosure, criminal conduct, and repeated basic security failures.
Microsoft’s November updates fixed more than 1,259 vulnerabilities yet disrupted printing, activation, installation, and SQL Server performance, underscoring the cost of unstable patching.
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.