Shared Responsibility Model
The Shared Responsibility Model is an agreement that divides security duties between a cloud service provider and its customer, making clear which protections each party is responsible for. In general, the provider secures the underlying cloud infrastructure, while the customer secures what they put in the cloud, such as their data, access controls, and applications. It helps organizations avoid dangerous assumptions that the provider handles everything automatically.
The Shared Responsibility Model is a security and compliance framework that delineates which cybersecurity processes and controls belong to the cloud service provider (CSP) versus the customer for a given cloud deployment. Providers such as AWS and Azure typically describe it as securing the infrastructure of the cloud itself, while customers remain responsible for security in the cloud, with the exact boundary varying by service model (IaaS, PaaS, or SaaS). From a governance standpoint, a virtual CISO should note that this model reallocates operational security tasks but does not transfer organizational or regulatory accountability away from the customer; even where the CSP contractually assumes control of certain layers, legal accountability for the customer's data, configurations, and compliance posture generally remains with the client organization and its officers. Common expert corrections include not treating the CSP as responsible for customer misconfigurations, identity and access management, or data classification simply because those functions run within the provider's platform.
Why it matters
The Shared Responsibility Model matters because one of the most common and costly sources of cloud security failures is the mistaken assumption that moving to a cloud provider transfers all security duties to that provider. In practice, providers such as AWS and Azure describe themselves as responsible for the security of the cloud infrastructure itself, while the customer remains responsible for security in the cloud, including data, access controls, and application configuration. When organizations do not understand where that boundary sits, gaps emerge in exactly the areas the provider never agreed to cover, such as identity and access management, data classification, and workload configuration.
For security leaders and the executives they advise, the model is as much a governance concern as a technical one. As AWS describes it, the shared model can help relieve the customer's operational burden, but a virtual CISO should be clear that reallocating operational tasks is not the same as transferring accountability. Even where a cloud service provider contractually assumes control of certain layers, legal and regulatory accountability for the customer's data, configurations, and overall compliance posture generally remains with the client organization and its officers. Treating the provider as accountable for a customer's own misconfiguration is a mistake an experienced practitioner would insist on correcting.
The boundary is also not fixed. It shifts depending on whether a workload is delivered as IaaS, PaaS, or SaaS, meaning the customer's obligations expand or contract with each service model. Because the exact division varies by provider and by service, the value of understanding the model depends heavily on organizations reading provider documentation carefully and mapping their own controls against it rather than assuming a single universal split applies everywhere.
Who it's relevant to
Inside SRM
Common questions
Answers to the questions practitioners most commonly ask about SRM.