Stellar Cyber's AI Triage Slashes Analyst Workload, Saves 19 Minutes Per Hour
Stellar Cyber's new Agentic Auto Triage tool significantly reduces analyst workload by automatically resolving 64% of security alerts, saving an average of 19 minutes per hour per analyst.
Stellar Cyber has unveiled promising results from customer trials of its Agentic Auto Triage tool, demonstrating a substantial reduction in the workload for security analysts. During a 124-day trial period within customer environments, the system processed 138,475 alerts, successfully identifying and closing 64% of them—amounting to 8,047 tickets—as confident false positives. This automation allows analysts to focus on more critical threats, reclaiming valuable time.
Beyond false positives, the Agentic Auto Triage tool escalated 1,875 alerts as genuine threats requiring human intervention, while logging others as informational. The cumulative effect of this automated triage is a reported saving of approximately 19 minutes per hour for each analyst. Stellar Cyber highlights that this reclaimed time can be reinvested in more proactive security measures and complex investigations, effectively allowing junior analysts to operate at a higher skill level and helping organizations combat the industry-wide hiring squeeze.
In an industry flooded with AI-driven solutions, Stellar Cyber emphasizes a pragmatic approach to its own "autonomous AI" claims. The company encourages potential buyers to scrutinize vendor claims and points to the tangible results from its customer trials as proof of efficacy. Lisa Liu, communications and corporate marketing manager at Stellar Cyber, stated that the true value lies in tools that "act symbiotically with your analysts and boost their productivity," rather than unsupervised agents that could introduce new risks.
Crucially, the trials revealed a remarkable 99.7% agreement rate between human analysts and the AI agent on alert verdicts. Despite this high accuracy, Stellar Cyber maintains a human review step in the workflow, deeming it essential for "the playbook for the foreseeable future." This cautious approach acknowledges the potential for even sophisticated AI to make errors, and Liu stressed that without this oversight, "assets become liabilities" due to the speed at which AI operates.
The persistent challenge of alert noise, originating from detection layers, is a significant pain point for the cybersecurity industry. Stellar Cyber's solution addresses this by absorbing the high volume of alerts at machine speed. Liu anticipates this volume will only increase as adversaries adopt more sophisticated AI tools, making effective alert management critical for organizational success.
Stellar Cyber's Multi-Layer AI architecture underpins this automation, stacking data transformation, detection management, and remediation. The guiding principle, according to Liu, is "full visibility," ensuring that even at the most automated levels, human analysts retain agency and the ability to intervene. This human-augmentation approach empowers analysts to enhance their own capabilities and wield AI tools more effectively, fostering a collaborative environment between humans and machines.
Organizations looking to understand how AI can genuinely augment their security operations and reduce alert fatigue can learn more about Stellar Cyber's approach. The company suggests that the burden of proof for AI claims should include a clear explanation of how the technology works in concert with human analysts, rather than relying solely on unsubstantiated efficiency figures.
Stellar Cyber will be showcasing its technology at Black Hat USA 2026, inviting attendees to visit booth #5542 for further demonstrations and discussions on AI-driven security automation.