Security | Threat Detection | Cyberattacks | DevSecOps | Compliance

Evaluating AI systems at Corelight

AI system evaluation is the process of continuously assessing AI system capabilities, limitations, and performance through quantitative and qualitative measures. Across the system lifecycle, evals provide continuous assurance: Validating system behavior before deployment and detecting drift, bias, and reliability issues in production.

Episode 21 - Building AI Harnesses to Unify Detection and Response

Corelight Senior Security Engineer Jordan Hair joins Richard Bejtlich to break down how defense teams can leverage agentic AI harnesses to transform traditional security operations. By wrapping deterministic code around large language models, Hare created automated agents for alert triage, threat hunting, and detection engineering that shrink routine investigations from 45 minutes down to seconds.

Securing Black Hat's NOC: Lessons from James Pope of Corelight

Corelight’s James Pope joins Dark Reading’s Joan Goodchild at Black Hat USA to share lessons from more than a decade defending one of cybersecurity’s most unique network environments: the Black Hat Network Operations Center (NOC). As SOC lead for the Black Hat NOC since 2014, James helps oversee more than 100 analysts, threat hunters, and partners tasked with distinguishing legitimate security research from real attacks across a network built from scratch for the conference.

Mythos: When Al becomes the attacker, the network becomes the first line of defense.

Cyber defense in the age of Mythos Advanced AI has fundamentally shifted the security landscape, shrinking the window for vulnerability exploitation from weeks to hours. When standard patching workflows can't keep pace, your network becomes your most critical line of defense. In this video, we explore how Corelight transforms network traffic into actionable security insights to power your SOC. The best data drives the best defense. Discover how to improve your SOC outcomes by up to 300% over legacy data.