Microsegmentation has been one of the most repeated promises of Zero Trust for years, but also among the most challenging to implement in practice. On paper, isolating applications, limiting lateral movement, and reducing the attack surface seem like obvious decisions. However, in daily operations, many organizations remain held back by fears of service disruption, the complexity of mapping dependencies, and a lack of staff to translate designs into actual policies. That’s where Akamai aims to intervene now, introducing a new set of AI capabilities for Guardicore Segmentation.
The company announced on March 24 new features for its Akamai Guardicore Segmentation platform with a very specific goal: leveraging Artificial Intelligence to identify, analyze, and interpret application behavior, and from there, generate ready-to-apply policies. Akamai claims this approach can accelerate segmentation projects, reduce manual effort, and enable companies to apply controls with greater confidence in hybrid, cloud, Kubernetes, and AI workloads.
The main challenge wasn’t understanding Zero Trust, but implementing it without breaking anything
One of the historical obstacles of microsegmentation has always been the same: knowing what communicates with what, the actual dependencies between applications, and what impact a policy will have when moving from observation mode to enforcement mode. Akamai states that it has analyzed over 500 segmentation projects to identify bottlenecks and has focused these improvements on resolving that specific phase where many initiatives slow down or remain incomplete.
The core of the announcement lies in several new capabilities. The first is continuous discovery, designed to provide real-time visibility and build a stronger foundation for Zero Trust. The second is an AI that “understands” applications—discovering behavior, proposing policies, explaining why they are generated, simulating impact, and validating if the environment is ready before activating controls. Additionally, there is the so-called “proof-driven enforcement,” continuous risk containment, and delegated workflows that allow application owners to participate directly in approvals and deployment.
This approach makes sense in a context where attackers are moving increasingly faster within networks once they gain initial access. Akamai has long positioned Guardicore Segmentation as a tool to curb lateral movement, reduce ransomware impact, protect critical assets, and reinforce cloud migrations with granular controls. The innovation now is to reduce the manual component of this work and transfer part of the interpretation and policy design effort to an AI layer.
Less console, more context and automation
Akamai’s messaging revolves around a clear idea: microsegmentation cannot continue relying on lengthy projects, intensive consulting, and endless validation if it aims to become a more widespread practice. That’s why it emphasizes that these new functions are not just about recommending policies but also maintaining continuous visibility, validating readiness before enforcement, and fueling policy actions with exposure and risk analyses.
This detail is important because a segment of the market has historically seen segmentation as desirable but too costly to maintain. Akamai seeks to change this perception with an approach where AI does not replace security logic but helps translate telemetry and observed behavior into actionable policies. It also introduces a significant organizational component: a portal for application owners, aimed at involving those who understand the actual functioning of services more directly and speeding up approvals without turning the entire process into an ongoing tug-of-war between security and operations.
Akamai presents these enhancements as particularly useful for organizations with hybrid environments, cloud workloads, Kubernetes, and AI applications that need to reduce lateral movement risk, limit incident impact scope, and meet increasing audit, compliance, and data sovereignty requirements. This orientation aligns with a widespread business reality: the more distributed the infrastructure, the harder it is to maintain uniform controls, and the more valuable a platform that continuously understands dependencies becomes.
A response to operational fears that have hindered many projects
The real question is whether this AI layer can resolve the most common fear in segmentation projects: applying a policy and causing an unexpected disruption. Akamai claims that its new features allow simulation of impact, explanation of policies, and environment validation before enforcement, which should lower that risk. In theory, this moves us from a model where companies “guess” what to block to one where they have more evidence before acting.
This does not mean segmentation becomes fully automatic and unsupervised. The quality of the results will still depend on inventory accuracy, traffic observation, the involvement of application teams, and how disciplined the review of actual system behavior is. However, it suggests a significant shift: the market is no longer just selling microsegmentation as a desirable architecture but as a practice that should be faster, more demonstrable, and less dependent on scarce specialists.
At a time when Zero Trust has nearly become an aspirational standard for large organizations, the main barrier is no longer just talk but execution. Akamai believes AI can help close that gap in Guardicore Segmentation. The market’s response remains to be seen, but the message is clear: the next phase of Zero Trust will not only focus on identity and access but also on better segmentation—more contextually aware and with less operational friction.
Frequently Asked Questions
What has Akamai announced for Guardicore Segmentation?
Akamai has introduced new AI-powered capabilities to find application behavior, generate segmentation policies, simulate their impact, and validate their application before enforcement.
What problem does this update aim to solve?
It aims to reduce the complexity and fear of implementing microsegmentation policies that might affect services by providing continuous discovery, policy generation, and pre-validation supported by AI.
Which environments is Akamai Guardicore Segmentation designed for?
It is targeted at organizations with hybrid infrastructure, cloud, Kubernetes, and AI workloads that need to reduce lateral movement, ransomware impact, and compliance challenges.
What does “proof-driven enforcement” mean in this context?
According to Akamai, it refers to a policy enforcement approach based on evidence and prior validation, aiming to reduce attack surfaces and scale Zero Trust without increasing staff or assuming excessive operational risks.
via: akamai