CVE-2025-32463 lets any local user reach root through sudo’s chroot option, and it patches in an afternoon. The detection most teams wrote for it alerts on the wrong binary — here’s the field to key on and the false positives that flood you first.
Ubuntu 24.04 enabled AppArmor mediation of unprivileged user namespaces by default, then Qualys published three ways around it. Here’s what the control actually stops, the audit chain that proves it fired, and how to detect abuse without flooding the SOC.
An out-of-bounds read in Ollama’s model quantization path leaks process heap, and the data leaves through Ollama’s own /api/push. The detection problem isn’t the read, it’s that the box almost certainly isn’t logging anything.
Device code phishing produces a clean, MFA-satisfied sign-in on Microsoft’s own infrastructure — and most of the telemetry that betrays it sits in the Entra non-interactive log table teams drop to save money. Here’s where the detection actually lives, how the threshold flips between a Windows shop and a dev-heavy tenant, and the persistence artifacts the closeout always skips.
GTG-1002 showed an AI agent running recon through exfiltration at machine speed across roughly 30 targets. A blue-team analysis of the behavioral tells, the identity and SIEM signals that expose autonomous operations, how to break the adversary’s loop, and where defensive AI agents help versus where the human-in-the-loop line stays.
Encryption-type detancy deton Kerberos TGS requests still catches lazy roasting, but as RC4 gets disabled the durable signal moves to behavior. What to grep for, why the AES downgrade tell degrades, and how to kill service-account false positives before they bury the SOC.
Most teams write their first ClickFix detection against the Windows Run dialog’s RunMRU artifact. FileFix and the App-V proxy variant route around it the next day. Here is the detection that survives, and what the first week of tuning has to fix.
Indirect prompt injection is hard to catch in the prompt and easy to miss in the model output. The reliable detection surface is the outbound fetch the rendered answer triggers — here is how to instrument it and where the false positives come from.
Bring-your-own-vulnerable-driver EDR killers are a standard stage in ransomware intrusions now, not an exotic capability. The driver blocklist is the weakest control you have against them. Here is what to instrument instead.
A clean netstat is not proof of safety: eBPF backdoors like LinkPro keep an internal listener, rewrite ports through XDP/TC, and can make bpftool lie about themselves. The durable signal is the bpf() syscall at load time, and on Cilium-heavy fleets, telling real loaders from noise is most of the work.
