AI is now part of almost every conversation in security operations. Most teams are already investing in it, experimenting with it, or trying to understand where it fits. The challenge is not whether to adopt AI, but how to apply it in a way that actually improves outcomes.

At the Rapid7 Global Cybersecurity Summit, the session The AI Dilemma: Automating Defense Without Surrendering Judgment explores how AI is being used in the SOC today, and where it creates real value in practice.

The discussion centers on a set of assumptions that often shape how teams approach AI, and why those assumptions do not always hold up in real environments.

Myth 1: AI will replace analysts

Across the session, there is a consistent focus on how AI supports investigation workflows by reducing repetitive work and surfacing relevant context, allowing analysts to focus on decisions that require judgment. AI helps teams move faster, but responsibility and accountability still sit with people. TL;DR, the role of the analyst is evolving, but it is not disappearing.

Myth 2: More automation means better security outcomes

Automation is valuable when it is applied in the right places. In practice, teams are finding the most benefit in areas such as enrichment, summarization, and triage, where large volumes of data need to be processed quickly. High-impact actions such as containment or configuration changes still require oversight, particularly when they can affect production systems or business operations.

Myth 3: Speed is more important than transparency

As adoption increases, trust becomes more important and analysts need to understand how a conclusion was reached before they act on it, especially in high-pressure situations. The session highlights how explainability builds confidence over time, allowing teams to rely on AI outputs without losing control of the decision-making process.

Myth 4: AI is only about efficiency gains

Efficiency is part of the story, but the impact runs deeper. AI helps connect signals across fragmented environments, reduces cognitive load, and supports more consistent decision-making. It also changes how teams approach investigation by making it easier to surface patterns and identify relationships that would be difficult to see manually.

Myth 5: Attackers benefit more from AI than defenders

Both attackers and defenders are learning how to use AI, and both are moving quickly. What matters for security teams is how they apply it within their own workflows. The session explores how AI strengthens detection, investigation, and response when it is integrated into existing processes rather than treated as a standalone capability.

Where AI creates real value in the SOC

Across the discussion, a clear pattern emerges. AI delivers the most value when it is applied to high-volume, context-heavy tasks, where it can process data, highlight signals, and recommend next steps. Analysts remain central to interpreting those signals, understanding intent, and deciding how to respond.

This balance between automation and oversight is what allows teams to scale their operations without losing confidence in their decisions. It also reflects how AI is being adopted across the industry, with most organizations maintaining moderate to high levels of human involvement as they build trust in these systems.

For SOC leaders, practitioners, and teams exploring AI, the session offers a grounded view of how these technologies are being applied today, and how that approach is continuing to evolve.

Watch the full session to explore how transparent AI supports better decisions in the SOC and how teams are applying it in practice.