Blog Post

Microsoft Entra Blog
4 MIN READ

End VPN gaps with identity-first access

Janice_Ricketts's avatar
Aug 05, 2026

Protect AI, SaaS, private apps, and web traffic with adaptive policies that respond to risk in real time.

Traditional VPNs extend network access—but they were not built continuously evaluate identity, device posture, location, user risk, or session context. That gap can leave AI apps, SaaS, on-premises applications, unmanaged services, and internet traffic unevenly protected.

Zero Trust closes this gap by putting identity and policy at the center of every access decision. Microsoft principles are simple: verify explicitly, use least-privilege access, and assume breach. As user, device, application, location, and risk signals change, access decisions can change with them.

Conditional Access serves as the policy engine, while Global Secure Access—Microsoft Entra Internet Access and Microsoft Entra Private Access—extends identity-driven controls across AI-enabled apps, SaaS, on-premises resources, and internet sessions without the complexity of legacy VPN architectures.

What Zero Trust means here

In this scenario, Zero Trust means evaluating every access request using identity, device, application, location, and risk signals; granting only the minimum access required; and adjusting access when posture or risk changes. It is not about putting a harder shell around the network. It is about moving policy enforcement closer to the user, session, and resource.

TL;DR: Your quick guide

  • Apply Zero Trust principles everywhere, not just to Microsoft 365 or core SaaS apps.
  • Use the same identity-driven model across AI apps, on-premises apps, SaaS, and internet traffic.
  • Extend risk-based Conditional Access to protect more than cloud apps.
  • Use Microsoft Entra Internet Access for threat-aware browsing, web content filtering, and granular SaaS controls.

Why Zero Trust must apply to every resource

Most organizations have invested heavily in cloud security, but attackers still seek out the weakest layer. This often means targeting new AI experiences that are not yet governed with the same rigor as core business apps, as well as on-premises applications, unmanaged SaaS, and internet sessions. Traditional VPNs and static firewalls struggle in this environment because they do not natively adapt to real-time risk signals or the realities of hybrid work.

Zero Trust enforcement addresses this gap by using explicit verification and least-privilege access instead of broad, network-based trust. In Microsoft’s model, Conditional Access brings together signals such as identity, device state, location, target application, and risk to make policy decisions. When applied consistently across AI apps, SaaS, on-premises apps, and internet access, it helps limit lateral movement, reduce attacker dwell time, and simplify security operations.

The goal is not to build a separate policy model for each type of resource. The goal is to establish one approach that can be applied broadly:

  • Verify explicitly before and during access.
  • Limit access to only what is needed.
  • Use risk and session context to adapt enforcement.
  • Keep policies understandable enough to scale across different app types.

Business value for large enterprises

The value of this model is not only technical. It changes how teams operate.

  • Lower breach risk through automated, policy-driven responses to compromise signals.
  • Faster operations by reducing VPN complexity and moving toward a cloud-native Security Service Edge model.
  • Improved user experience through secure access that avoids split-tunnel friction and inconsistent connectivity.
  • Better cost efficiency by reducing dependence on on-premises hardware and overlapping security tools.

Just as important, teams gain a more reusable security model. Instead of creating one set of rules for AI tools, another for on-premises apps, and another for SaaS, they can extend a common Zero Trust framework across all of them.

In practice, this means that a compromised account can trigger a faster, policy-driven response across both cloud and private-resource access paths instead of relying on manual containment.

Key takeaways

  • Extend Zero Trust beyond the cloud and apply it to every app, session, and resource.
  • Use Microsoft Entra risk detections and automated remediation to trigger real-time threat responses.
  • Extend enforcement to internet traffic, nonfederated apps, and AI-driven experiences without adding another layer of tools.
  • Reduce complexity and cost by retiring VPN-heavy architectures in favor of policy-based access.

The path forward

Global Secure Access, together with risk-based Conditional Access, gives enterprises a clearer path to applying one identity-driven access model across AI applications, on-premises applications, SaaS, and internet traffic. Through Microsoft Entra Internet Access and Microsoft Entra Private Access, Global Secure Access helps reduce policy silos, improve responses to user risk, and strengthen Zero Trust across hybrid operations.

For security leaders, the opportunity is not simply to modernize remote access. It is to build an access strategy that can continuously evaluate trust, respond to changing risk, and apply the same Zero Trust principles consistently, regardless of the type of application or resource users need to access.

 

Janice Ricketts

Janice Ricketts, MBA, BSEE, PMP | LinkedIn

 

Additional resources

 Learn more about Microsoft Entra

Prevent identity attacks, ensure least-privilege access, unify access controls, and improve the user experience with comprehensive identity and network access solutions across on-premises environments and clouds.

 

Updated Jul 31, 2026
Version 1.0
No CommentsBe the first to comment