IAC-29.1: Real-Time Access Decisions

There is no evidence of a capability to utilize Machine Learning (ML) to make real-time access decisions based on advanced network analytics that leverages enterprise-wide data sources.

C|P-CMM1 is N/A, since a structured process is required to utilize Machine Learning (ML) to make real-time access decisions based on advanced network analytics that leverages enterprise-wide data sources.

Identification & Authentication (IAC) efforts are requirements-driven and governed at a local/regional level, but are not consistent across the organization. CMM Level 2 control maturity would reasonably expect all, or at least most, the following criteria to exist: o Implement and maintain an Identity & Access Management (IAM) capability for all users to implement “least privileges” Role Based Access Control (RBAC) practices for the management of user, group and system accounts, including privileged accounts. o Govern IAM technologies via RBAC to prohibit privileged access by non-organizational users, unless there is an explicit support contract for privileged IT support services.

• Logical Access Control (LAC) is decentralized (e.g., a localized/regionalized function) and uses non-standardized methods to implement secure, resilient and compliant practices.
• IT/cybersecurity personnel identify cybersecurity and data protection controls that are appropriate to address applicable statutory, regulatory and contractual requirements for logical access control.
• IT personnel:
• Active Directory (AD), or a similar technology, is primarily used to centrally manage identities and permissions with RBAC. Due to technical or business limitations, asset/process owners are empowered to operate a decentralized access control program for their specific systems, applications and services that cannot be integrated into AD.
• IAM controls are primarily administrative in nature (e.g., policies & standards) to manage accounts and permissions.
• Administrative processes exist to collect, validate and verify identity evidence of a user.

Identification & Authentication (IAC) efforts are standardized across the organization and centrally managed, where technically feasible, to ensure consistency. CMM Level 3 control maturity would reasonably expect all, or at least most, the following criteria to exist:

• An Identity & Access Management (IAM) function, or similar function, centrally manages permissions and implements “least privileges” Role Based Access Control (RBAC) practices for the management of user, group and system accounts, including privileged accounts.
• The Human Resources (HR) department governs personnel management operations and notifies IAM personnel of personnel role changes for RBAC-based provisioning and deprovisioning actions.
• An IT Asset Management (ITAM) function, or similar function, categorizes endpoint devices according to the data the asset stores, transmits and/ or processes and applies the appropriate technology controls to protect the asset and data that conform to industry-recognized standards for hardening (e.g., DISA STIGs, CIS Benchmarks or OEM security guides).
• An IT infrastructure team, or similar function, ensures that statutory, regulatory and contractual cybersecurity and data privacy obligations are addressed to ensure secure configurations are designed, built and maintained.
• Active Directory (AD), or a similar technology, is used to centrally manage identities and permissions. Only by exception due to a technical or business limitation are solutions authorized to operate a decentralized access control program for systems, applications and services.
• Administrative processes and technologies enforce Attribute-Based Access Control (ABAC) to enable policy-driven, dynamic authorizations and support secure sharing of information

See C|P-CMM3. There are no defined C|P-CMM4 criteria, since it is reasonable to assume a quantitatively-controlled process is not necessary to utilize Machine Learning (ML) to make real-time access decisions based on advanced network analytics that leverages enterprise-wide data sources.

See C|P-CMM4. There are no defined C|P-CMM5 criteria, since it is reasonable to assume a continuously-improving process is not necessary to utilize Machine Learning (ML) to make real-time access decisions based on advanced network analytics that leverages enterprise-wide data sources.