Agentic security: Detection and response at machine speed

<p>After talking with enterprise security leaders over the past year, one thing has become clear: the rise of autonomous AI agents is the most significant shift in security posture since the move to cloud. Organizations across every industry are adopting AI agents that authenticate on behalf of users, execute multistep workflows, and make decisions across infrastructure, often without waiting for human approval. Security operations need to keep pace.</p>

<p>At <span class="LinkEnhancement"><a class="Link" href="https://aws.amazon.com" target="_blank" rel="noopener" data-cms-ai="0">Amazon Web Services (AWS)</a></span>, we believe security should evolve ahead of AI adoption, not behind it. That belief drove our team to collaborate with the SANS Institute on a new chapter in the <span class="LinkEnhancement"><a class="Link" href="https://www.sans.org/webcasts/sans-2026-cloud-security-exchange" target="_blank" rel="noopener" data-cms-ai="0">2026 Cloud Security Exchange eBook</a></span>, where we lay out a practical framework for securing agentic workloads at enterprise scale.</p>

<h2><b>The challenge: Threats now move at machine speed</b></h2>

</div>

<p>Traditional security was built for deterministic systems with predictable inputs and outputs. Agentic workloads break those assumptions. The same prompt can produce a compliant response on one request and a policy-violating response on the next. Agents adapt their behavior over time as they interact with users, data, and tools and operate with genuine autonomy: connecting to APIs, chaining actions together, and making independent decisions.</p>

<p>These properties mean that security controls designed for one-time assessments no longer suffice. Detection and response need to operate continuously and at machine speed.</p>

<p>What makes this urgent is the gap between adoption velocity and security maturity. Although 80% of organizations have adopted AI, only 10% govern it. Agents are being built by an expanding population of developers—including those using low-code tools—creating governance challenges that existing security programs must be extended to address.</p>

<h2><b>Extending what already works</b></h2>

</div>

<p>The good news, agentic security isn’t a blank slate. It builds on the same principles security teams already apply: identity governance, least privilege, defense in depth, and backup and recovery. What changes is how those principles are implemented when workloads are autonomous and probabilistic. In our eBook chapter, we cover four foundational areas:</p>

<li><b>Agent identity and governance:</b> Every agent needs its own identity with temporary, scoped credentials rather than persistent, broad access. This extends <span class="LinkEnhancement"><a class="Link" href="https://aws.amazon.com/security/zero-trust/" target="_blank" rel="noopener" data-cms-ai="0">zero trust principles</a></span> to AI agents, where every request is authenticated and authorized independently, and every action has a traceable authorization chain. When a single agent combines access to sensitive data, the ability to communicate externally, and exposure to untrusted content, the risk profile changes significantly. Design patterns that prevent any single component from combining all three reduce that risk substantially.</li>

<li><b>Evolving detection for agentic workloads:</b> Static, rule-based detection designed for human activity patterns can’t keep up with agent behavior. Organizations need continuous behavioral monitoring, living baselines that adapt as agents evolve, and instrumented observation that surfaces anomalies in real time. <span class="LinkEnhancement"><a class="Link" href="https://aws.amazon.com/guardduty/" target="_blank" rel="noopener" data-cms-ai="0">Amazon GuardDuty</a></span> delivers this today, analyzing security signals continuously to detect threats as they emerge.</li>

<li><b>Response that balances speed with precision:</b> When threats move at machine speed, response must be automated and tiered: some agent behaviors should be contained immediately, others require human judgment. The response framework we outline distinguishes between actions that can be automated safely and those that need escalation.</li>

<li><b>From single agents to multiagent ecosystems:</b> Agents are already composing into teams, delegating subtasks, negotiating access, and coordinating across organizational boundaries. Each stage of this evolution inherits every security requirement that came b