-
The Agentic SOC That Refuses to Guess
A large US-based MSSP ran five adversarial tests against Morpheus APD 2.0’s AI SOC guardrails. It passed all five with zero false all-clear verdicts. When…
-
Two Tools, Two Bills: The Consolidation Hiding In Your SOC Budget
Most SOCs pay two bills, a SOAR and an alert-automation tool, to do one job. Here is the cleanest consolidation in the 2026 stack, and…
-
Cost Savings or Freed Capacity? Field Notes From Agentic SOC Deployments
Field notes from production agentic SOC deployments: where the returns land, what the no-analysts pitch skips, and the proof to demand from any vendor.
-
Control Evidence by Design: How To Oversee an AI You Can’t Watch in Real Time
You can’t watch an autonomous SOC in real time. Here’s the architecture that makes human oversight provable: read-only investigation, gated actions, one record.
-
Keep ServiceNow, Kill the Meter, Own the Models
Keep ServiceNow for IT, drop the Now Assist usage meter, and run a SOC AI you can audit. The practical case for moving security operations…
-
The 3 a.m. Question: Who’s Liable When Your AI Acts Alone?
When your AI acts in the SOC at 3 a.m., liability stays with you. Here’s how to prove what it did and defend autonomy in…
-
The Story of an Alert, in 8 Stages You Can Prove
Investigating every alert at L2 depth is an engineering problem. Here is the 8-stage lifecycle behind it: autonomous at every stage, governed at every stage.
-
What a Governed Agentic SOC Does When It Can’t Be Sure
The most useful moment in an agentic SOC demo is the failure path, not the clean verdict. Here’s what to ask to see, and what…
-
100 Wrong Verdicts a Day: The Fine Print Inside a “99% Accurate” AI SOC
LLMs perform pattern completion over whatever context they’re given. Why AI triage reads missing evidence as benign, and the guardrail that prevents it.
D3 Security
Select a term…