AI in the SOC: Building Trust and Reliability
Embed Security CEO and Co-Founder Seth Summersett and Co-Founder and CTO Dr. Jeffrey Johns joined Dark Reading at Black Hat USA to discuss what it takes to build AI that security teams can trust in the SOC. Read the full article on Dark Reading, or watch the full Dark Reading News Desk Interview below.
Building AI that works in the real-world SOC
In their conversation with Dark Reading’s Joan Goodchild, Seth and Jeff discuss the gap between impressive AI demos and reliable, production-ready systems. Highlights from the conversation include:
- Reliability is critical. It’s relatively easy to get answers from an LLM. Building an AI system that delivers consistent, dependable results is much harder.
- Security context matters. Foundational models weren’t trained on deep security data, which can lead to inconsistent results and make it harder for analysts to trust the output.
- Feedback loops are essential. AI needs to learn from an organization’s environment and feedback to become more reliable and relevant over time.
- AI can shift analysts toward higher-value work. As AI takes on more information gathering, analysts can spend more time verifying findings, making decisions, and responding to threats.
