{"id":20956,"date":"2026-02-17T18:07:21","date_gmt":"2026-02-17T15:07:21","guid":{"rendered":"https:\/\/oberig-it.com\/uncategorized\/why-were-acquiring-strongdm-bringing-continuous-authorization-to-the-delinea-platform\/"},"modified":"2026-02-27T18:08:32","modified_gmt":"2026-02-27T15:08:32","slug":"why-were-acquiring-strongdm-bringing-continuous-authorization-to-the-delinea-platform","status":"publish","type":"post","link":"https:\/\/oberig-it.com\/en\/news\/why-were-acquiring-strongdm-bringing-continuous-authorization-to-the-delinea-platform\/","title":{"rendered":"Why We’re Acquiring StrongDM: Bringing Continuous Authorization to the Delinea Platform"},"content":{"rendered":"

What you will learn<\/em>
\nStrongDM solves the problem of extending privileged access into environments where traditional session-based controls simply don’t fit in a way that aligns with our direction for the Delinea Platform.<\/em><\/p>\n

When my team evaluated StrongDM, we weren’t looking for another secrets manager or another vault. We were looking for technology that could extend privileged access into environments where traditional session-based controls simply don’t fit\u2014CI\/CD pipelines, Kubernetes clusters, ephemeral infrastructure, and increasingly, autonomous AI agents making thousands of access decisions without human intervention.<\/p>\n

StrongDM solves that problem in a way that aligns with our direction for the Delinea Platform. Here’s why this acquisition matters and what it will mean for our customers.<\/p>\n

The problem we’re solving<\/h4>\n

Traditional PAM was designed for humans logging into systems. You authenticate, establish a session, and maintain standing access until that session ends. That model works well for interactive use cases, and it’s not going away\u2014it remains foundational to how Delinea protects privileged access today.<\/p>\n

But modern infrastructure doesn’t wait for humans. A CI\/CD pipeline deploying to production at 2 AM doesn’t have someone available to approve access. A Kubernetes pod spinning up needs credentials immediately, and it’ll be gone in minutes. An AI agent executing a multi-step workflow might need to access databases, APIs, and cloud resources in rapid succession, each action requiring different permissions.<\/p>\n

Organizations have tried to solve this by over-provisioning service accounts, embedding long-lived credentials in code, or creating broad IAM roles that violate least privilege. These are security risks masquerading as operational necessities.<\/p>\n

StrongDM takes a different approach: continuous, JIT runtime authorization evaluated at the moment of action, not at session establishment.<\/p>\n

What makes StrongDM different<\/h4>\n

Authorization at the protocol level<\/h5>\n

Most access control systems make a binary decision at connection time: can this identity access this resource? Once you’re in, you’re in.<\/p>\n

StrongDM evaluates authorization continuously, down to the data inside the protocol. For a database connection, this means policies can govern not just whether you can connect, but which tables you can query, which rows match your authorization context, and which operations you can perform\u2014evaluated in real-time as each query executes.<\/p>\n

This granularity matters enormously for agentic AI. When an autonomous agent is executing a complex workflow, you don’t want to grant it broad database access and hope it behaves. You want each action the agent takes to be authorized independently, scoped to exactly what that specific operation requires. If the agent attempts to access data outside its authorized scope, the request fails immediately\u2014not after a post-hoc audit finds the violation.<\/p>\n

StrongDM’s Cedar-based policy engine makes this practical. Cedar, developed by AWS and used in their Verified Permissions service, provides a policy language that’s both expressive enough for complex authorization logic and fast enough to evaluate inline without adding latency to every request. Policies are defined declaratively, version-controlled, and testable\u2014which matters when you’re governing thousands of automated decisions per minute.<\/p>\n

Built for engineers, not against them<\/h4>\n

One thing that stood out in our evaluation: StrongDM was clearly built by people who understand how engineers actually work.<\/p>\n

There’s no heavyweight agent to deploy. CLI workflows that DevOps teams already use continue to work\u2014StrongDM integrates via lightweight proxies and SDK hooks that feel native to the toolchain. Engineers don’t need to learn a new workflow or wait for tickets to get access to do their jobs. They authenticate once, and authorization happens transparently based on what they’re actually trying to do.<\/p>\n

For IT administrators, policy management is centralized but not bureaucratic. You define authorization policies in Cedar, attach them to identity contexts, and they apply consistently whether the access is coming from a human engineer, a CI\/CD pipeline, or an AI agent. When something doesn’t work, the audit trail tells you exactly why\u2014which policy denied the request, what context was evaluated, and what would need to change.<\/p>\n

This ease of use isn’t a nice-to-have. It’s essential for adoption. Security tools that create friction get bypassed. We’ve seen it countless times: teams embedding credentials in environment variables, sharing service accounts, or creating overly permissive IAM roles because the “secure” path was too slow. StrongDM removes that excuse by making the secure path the easy path.<\/p>\n

Unified visibility through the identity graph<\/h4>\n

StrongDM’s authorization decisions will feed directly into Delinea’s identity graph, providing centralized visibility into access patterns across your entire estate.<\/p>\n

This means security teams will get a unified view of who\u2014and what\u2014is accessing privileged resources, whether that’s a human administrator using Secret Server, a DevOps engineer accessing cloud infrastructure through StrongDM, or an AI agent executing automated workflows. You can answer questions like:<\/p>\n