Stellar Cyber’s Auto‑Triage AI Matches Human Analysts 99.7 % of the Time in Independent Study
What Happened — An independent 124‑day study of 138,475 real security alerts from Stellar Cyber customers found the platform’s Agentic Auto‑Triage capability produced the same verdict as human analysts 99.7 % of the time. The AI also reclaimed roughly 19 minutes of every analyst hour, equivalent to one full work‑day per week per analyst.
Why It Matters for Compliance & Audit Readiness
- Demonstrates that automated alert triage can be trusted to meet SOC 2 CC6.1 (Security Monitoring) requirements while providing continuous, auditable evidence of detection decisions.
- Reduces false‑positive noise, enabling teams to focus on high‑value investigations and maintain a defensible incident‑response timeline (MTTD/MTTR) that auditors can verify.
- Aligns with the “control‑mapping” capability: Verisq can ingest AI‑driven triage logs as proof that detection controls are operating effectively, simplifying evidence collection for SOC 2 audits.
Who Is Affected – SaaS security platforms, MSSPs, and any organization that runs a Security Operations Center (SOC) across technology, finance, healthcare, and retail sectors.
Recommended Actions
- Map the Auto‑Triage decision logs to SOC 2 CC6.1 (Security Monitoring) and CC7.1 (Incident Response) controls in your compliance framework.
- Implement continuous evidence collection of AI‑generated alerts and analyst overrides to create a defensible audit trail.
- Validate AI performance against internal baselines and document the 99.7 % concordance as part of your SOC 2 readiness evidence.
Source: Help Net Security – Stellar Cyber Auto‑Triage Study
Technical Notes – The study evaluated 138,475 alerts across 124 days of real‑world customer trials. No specific CVEs or vulnerabilities were involved; the focus is on AI‑driven alert correlation, prioritization, and false‑positive reduction. Source: same as above