Akamai adds AI features to Guardicore segmentation

Akamai has added AI-based features to its Guardicore Segmentation product to help organisations limit lateral movement after a breach.

The update focuses on how security teams create and enforce segmentation policies across hybrid IT, cloud, Kubernetes and AI workloads. Akamai says the product now uses AI to identify and interpret application behaviour, generate policies ready for enforcement, and test the likely effect of those policies before they are applied.

Segmentation restricts how attackers move through networks after gaining access. It has become a more prominent part of cyber defence as ransomware groups and other attackers seek to spread quickly across connected systems, increasing disruption and recovery costs.

One element of the update is continuous discovery, which gives security teams real-time visibility into application behaviour. That visibility can support Zero Trust security models, where users, systems and workloads are treated as untrusted by default and access is tightly controlled.

Another part of the release is AI-driven policy generation. Akamai says the system can automatically discover application behaviour, generate and explain segmentation rules at scale, simulate their impact, and validate whether they are ready for enforcement. The company is positioning the process as a way to reduce manual work for security teams and lower the risk of applying incorrect controls.

The product also adds what Akamai describes as continuous risk containment. This combines exposure-aware detection with segmentation enforcement so AI-generated findings can feed directly into policy actions. The aim is to reduce the blast radius of an attack by tightening controls more quickly when risks are identified.

A further change is delegated workflows for application owners through an App Owner Portal. This is intended to involve application owners more directly in approving segmentation changes, which can be a sticking point in large organisations where security teams need operational sign-off before rules are enforced.

Project lessons

Akamai linked the update to analysis of customer deployments, saying the new functions were informed by a review of more than 500 segmentation projects. The aim, it says, is to address common obstacles in technology, process and organisational behaviour.

"We analyzed more than 500 segmentation projects to pinpoint common bottlenecks and identify what drives success across technology, processes, and human behavior," said Ofer Wolf, Senior Vice President and General Manager of Enterprise Security at Akamai. "Those insights guided the development of these enhancements to Akamai Guardicore Segmentation that apply AI-powered segmentation throughout the entire risk containment lifecycle - from continuous discovery and intelligent insight to runtime assurance, exposure analysis, and rapid response."

The announcement comes as companies face growing pressure to show they can contain cyber incidents, not just prevent them. Boards, regulators and insurers are increasingly focused on resilience measures that can limit operational fallout when an attacker bypasses perimeter defences.

That pressure is especially acute in environments spanning on-premises infrastructure and multiple cloud services. In these settings, security teams often struggle to map application dependencies accurately enough to enforce segmentation without disrupting legitimate traffic. As a result, many projects move slowly or remain only partially implemented.

By using AI to observe behaviour and draft rules, Akamai is seeking to address that long-standing implementation problem. It is also trying to give security teams more evidence before controls are turned on through simulation and readiness validation, reducing the reluctance that can come with enforcing segmentation in production systems.

The focus on containment reflects a broader shift in cyber security strategy. Rather than assuming every intrusion can be stopped at the edge, many organisations are investing in internal controls designed to compartmentalise systems and data so a compromise in one area does not lead to wider disruption.

Guardicore Segmentation is aimed at organisations seeking to reduce the risk of lateral movement and ransomware while meeting audit, compliance and data sovereignty requirements, according to Akamai. Those demands have become harder to manage as workloads move across a mix of data centres, public cloud platforms and containerised environments.

Akamai's message is that segmentation must move from a one-off project to a continuous process.