+Article 26 Obligations incombant aux déployeurs de systèmes d’IA à haut risque
---+Article 26 1.
---+Article 26 2.
---+Article 26 3.
---+Article 26 4.
---+Article 26 5.
---+Article 26 6.
---+Article 26 7.
---+Article 26 8.
---+Article 26 9.
---+Article 26 10.
---+Article 26 11.
---+Article 26 12.

Article 26 Obligations incombant aux déployeurs de systèmes d’IA à haut risque

-

1. Vue d’ensemble

,
Article 26 1.
Les déployeurs de systèmes d’IA à haut risque prennent des mesures techniques et organisationnelles appropriées afin de garantir qu’ils utilisent ces systèmes conformément aux notices d’utilisation accompagnant les systèmes, conformément aux paragraphes 3 et 6.
Article 26 2.
Les déployeurs confient le contrôle humain à des personnes physiques qui disposent des compétences, de la formation et de l’autorité nécessaires ainsi que du soutien nécessaire.
Article 26 3.
Les obligations énoncées aux paragraphes 1 et 2 sont sans préjudice des autres obligations du déployeur prévues par le droit de l’Union ou le droit national et de la faculté du déployeur d’organiser ses propres ressources et activités aux fins de la mise en œuvre des mesures de contrôle humain indiquées par le fournisseur.
Article 26 4.
Sans préjudice des paragraphes 1 et 2, pour autant que le déployeur exerce un contrôle sur les données d’entrée, il veille à ce que ces dernières soient pertinentes et suffisamment représentatives au regard de la destination du système d’IA à haut risque.
Article 26 5.
Les déployeurs surveillent le fonctionnement du système d’IA à haut risque sur la base de la notice d’utilisation et, le cas échéant, informent les fournisseurs conformément à l’article 72. Lorsque les déployeurs ont des raisons de considérer que l’utilisation du système d’IA à haut risque conformément à la notice d’utilisation pourrait conduire à ce que le système d’IA présente un risque au sens de l’article 79, paragraphe 1, ils en informent, sans retard injustifié, le fournisseur ou le distributeur ainsi que l’autorité de surveillance du marché concernée, et suspendent l’utilisation de ce système. Lorsque les déployeurs ont détecté un incident grave, ils informent également immédiatement d’abord le fournisseur, puis l’importateur ou le distributeur et les autorités de surveillance du marché concernées de cet incident. Si le déployeur n’est pas en mesure de joindre le fournisseur, l’article 73 s’applique mutatis mutandis. Cette obligation ne couvre pas les données opérationnelles sensibles des déployeurs de systèmes d’IA qui sont des autorités répressives.

Si les déployeurs sont des établissements financiers soumis à des exigences relatives à leur gouvernance, à leurs dispositifs ou à leurs processus internes prévues par la législation de l’Union sur les services financiers, la conformité avec les règles relatives à la gouvernance, aux dispositifs, aux processus et aux mécanismes internes prévues dans la législation sur les services financiers vaut respect de l’obligation de surveillance énoncée au premier alinéa.
Article 26 6.
Les déployeurs de systèmes d’IA à haut risque assurent la tenue des journaux générés automatiquement par ce système d’IA à haut risque dans la mesure où ces journaux se trouvent sous leur contrôle, pendant une période adaptée à la destination du système d’IA à haut risque, d’au moins six mois, sauf disposition contraire dans le droit de l’Union ou le droit national applicable, en particulier dans le droit de l’Union sur la protection des données à caractère personnel.

Si les déployeurs sont des établissements financiers soumis à des exigences relatives à leur gouvernance, à leurs dispositifs ou à leurs processus internes prévues par la législation de l’Union sur les services financiers, ils tiennent à jour les journaux dans le cadre de la documentation conservée en vertu de la législation pertinente de l’Union sur les services financiers.
Article 26 7.
Avant de mettre en service ou d’utiliser un système d’IA à haut risque sur le lieu de travail, les déployeurs qui sont des employeurs informent les représentants des travailleurs et les travailleurs concernés qu’ils seront soumis à l’utilisation du système d’IA à haut risque. Ces informations sont fournies, le cas échéant, conformément aux règles et procédures prévues par le droit de l’Union et le droit national et aux pratiques en matière d’information des travailleurs et de leurs représentants.
Article 26 8.
Les déployeurs de systèmes d’IA à haut risque qui sont des autorités publiques ou des institutions, organes ou organismes de l’Union, respectent les obligations en matière d’enregistrement prévues à l’article 49. Dans le cas où ces déployeurs constatent que le système d’IA à haut risque qu’ils envisagent d’utiliser n’a pas été enregistré dans la base de données de l’UE visée à l’article 71, ils n’utilisent pas ce système et informent le fournisseur ou le distributeur.
Article 26 9.
Le cas échéant, les déployeurs de systèmes d’IA à haut risque utilisent les informations fournies en application de l’article 13 du présent règlement pour se conformer à leur obligation de procéder à une analyse d’impact relative à la protection des données en vertu de l’article 35 du règlement (UE) 2016/679 ou de l’article 27 de la directive (UE) 2016/680.
Article 26 10.
Sans préjudice de la directive (UE) 2016/680, dans le cadre d’une enquête en vue de la recherche ciblée d’une personne soupçonnée d’avoir commis une infraction pénale ou condamnée pour avoir commis une infraction pénale, le déployeur d’un système d’IA à haut risque pour l’identification biométrique à distance a posteriori demande l’autorisation, ex ante ou sans retard injustifié et au plus tard dans les 48 heures, d’une autorité judiciaire ou administrative dont la décision est contraignante et soumise à un contrôle juridictionnel, pour l’utilisation de ce système, sauf lorsqu’il est utilisé pour l’identification initiale d’un suspect potentiel sur la base de faits objectifs et vérifiables directement liés à l’infraction. Chaque utilisation est limitée à ce qui est strictement nécessaire pour enquêter sur une infraction pénale spécifique.

Si l’autorisation demandée en application du premier alinéa est rejetée, l’utilisation du système d’identification biométrique à distance a posteriori lié à l’autorisation demandée est interrompue avec effet immédiat et les données à caractère personnel liées à l’utilisation du système d’IA à haut risque pour lequel l’autorisation a été demandée sont supprimées.

En aucun cas, ce système d’IA à haut risque pour l’identification biométrique à distance a posteriori ne peut être utilisé à des fins répressives de manière non ciblée, sans aucun lien avec une infraction pénale, une procédure pénale, une menace réelle et actuelle ou réelle et prévisible d’une infraction pénale, ou la recherche d’une personne disparue spécifique. Il convient d’assurer qu’aucune décision produisant des effets juridiques défavorables à l’égard d’une personne ne puisse être prise par les autorités répressives sur la seule base des sorties de tels systèmes d’identification biométrique à distance a posteriori.

Le présent paragraphe est sans préjudice de l’article 9 du règlement (UE) 2016/679 et de l’article 10 de la directive (UE) 2016/680 pour le traitement des données biométriques.

Indépendamment de la finalité ou du déployeur, chaque utilisation de ces systèmes d’IA à haut risque est documentée dans le dossier de police pertinent et est mise à la disposition de l’autorité de surveillance du marché concernée et de l’autorité nationale chargée de la protection des données sur demande, à l’exclusion de la divulgation de données opérationnelles sensibles liées aux services répressifs. Le présent alinéa est sans préjudice des pouvoirs conférés par la directive (UE) 2016/680 aux autorités de contrôle.

Les déployeurs soumettent aux autorités de surveillance du marché concernées et aux autorités nationales chargées de la protection des données des rapports annuels sur leur utilisation de systèmes d’identification biométrique à distance a posteriori, à l’exclusion de la divulgation de données opérationnelles sensibles liées aux services répressifs. Les rapports peuvent être agrégés pour couvrir plus d’un déploiement.

Les États membres peuvent adopter, conformément au droit de l’Union, des lois plus restrictives sur l’utilisation de systèmes d’identification biométrique à distance a posteriori.
Article 26 11.
Sans préjudice de l’article 50 du présent règlement, les déployeurs de systèmes d’IA à haut risque visés à l’annexe III, qui prennent des décisions ou facilitent les prises de décision concernant des personnes physiques, informent lesdites personnes physiques qu’elles sont soumises à l’utilisation du système d’IA à haut risque. Pour les systèmes d’IA à haut risque utilisés à des fins répressives, l’article 13 de la directive (UE) 2016/680 s’applique.
Article 26 12.
Les déployeurs coopèrent avec les autorités compétentes concernées à toute mesure prise par ces autorités à l’égard du système d’IA à haut risque en vue de mettre en œuvre le présent règlement.

1.1 Références

1.2 Exigences identifiées

1.3 Normes associées

2. Exigences identifiées

Exigences

3. Normes associées

Normes
SCF AI & Autonomous Technologies Deployment

Description

Mechanisms exist to ensure the deployment of Artificial Intelligence (AI) and Autonomous Technologies (AAT) includes appropriate technical and organizational measures so that AAT are used in accordance with the AAT developer-provided instructions.

Possible Solutions & Considerations

Micro-Small Business (<10 staff) / BLS Firm Size Classes 1-2

∙ Deploy AI tools only for documented, approved purposes

Small Business (10-49 staff) / BLS Firm Size Classes 3-4

∙ AI deployment approval process
∙ Document intended use and technical measures

Medium Business (50-249 staff) / BLS Firm Size Classes 5-6

∙ Formal AI deployment policy
∙ Technical and organizational measures for AI deployment

Large Business (250-999 staff) / BLS Firm Size Classes 7-8

∙ AI deployment governance program
∙ Formal approval workflows
∙ Post-deployment monitoring

Enterprise (> 1,000 staff) / BLS Firm Size Class 9

∙ Enterprise AI deployment governance framework
∙ Automated deployment controls
∙ Continuous monitoring
∙ Regulatory compliance

SCR-CMM

Level 0 Not Performed

Practices are non-existent, based on the inability to demonstrate an implemented and operational capability. A reasonable person would conclude the control is not being performed.

Level 1 Performed Informally

Artificial Intelligence and Autonomous Technology (AAT) domain capabilities are ad hoc and inconsistent. Capability criteria associated with this control may include:
▪ Policies, standards & procedures associated with AAT domain capabilities provide limited coverage due to the depth and breadth of the existing documentation.
▪ AAT-related processes are expected to follow the organization's existing processes (e.g., incident response, asset management, change control and risk assessments).
▪ No formal Governance, Risk & Compliance (GRC) team exists to provide AAT oversight, where the Chief Information Officer (CIO), or similar function, governs technology decisions what is acceptable for AAT within the organization.

Level 2 Planned Tracked

Artificial Intelligence and Autonomous Technology (AAT) capabilities are requirements-driven, but are not standardized across the entity (e.g., local/regional level consistency). Capability criteria associated with this control reasonably expect the following criteria to exist:
▪ Policies and standards associated with AAT domain capabilities are formally documented and centrally-managed by the entity.
▪ Standardized Operating Procedures (SOP) associated with AAT domain capabilities are documented and maintained by process owners.
▪ IT and/or cybersecurity personnel work with business stakeholders and process owners to appropriately scope and reasonably implement cybersecurity and data protection controls associated with AAT domain capabilities to address applicable statutory, regulatory and/or contractual requirements for Technology Assets, Applications, Services and/or Data (TAASD).
▪ Artificial Intelligence (AI)-related controls are primarily administrative and preventative in nature (e.g., policies, standards and procedures & guidelines).
▪ Asset management may be a defined function (e.g., team or department) or assigned as an additional duty to existing IT and/or cybersecurity personnel.
▪ AAT is regarded as a technology and governed by the entity's existing IT governance practices.
▪ No formal Governance, Risk & Compliance (GRC) team exists to provide oversight of AAT-related activities. GRC functions are assigned to existing IT and/or cybersecurity personnel.

Level 3 Well Defined

Artificial Intelligence and Autonomous Technology (AAT) capabilities are standardized across the entity for applicability to People, Processes, Technologies, Data and/or Facilities (PPTDF) to ensure consistency for Technology Assets, Applications, Services and/or Data (TAASD). Capability criteria associated with this control reasonably expect the following criteria to exist:
▪ Policies and standards associated with AAT domain capabilities are formally documented and centrally-managed by the entity's Governance, Risk & Compliance (GRC) team, or similar function.
▪ Standardized Operating Procedures (SOP) associated with AAT domain capabilities are well-documented and kept current by process owners.
▪ An Artificial Intelligence Governance (AIG) team, or similar function, is appropriately staffed and supported to implement and maintain AAT domain capabilities.
▪ Technology is leveraged to enhance the efficiency and accuracy of AI governance, risk management and compliance operations (e.g., dedicated AI governance platform).
▪ The entity's Governance, Risk & Compliance (GRC) team, or similar function, works with business stakeholders and process owners to appropriately scope and reasonably implement cybersecurity and data protection controls associated with AAT domain capabilities to address Minimum Compliance Requirements (MCR) (e.g., applicable statutory, regulatory and/or contractual requirements) and Discretionary Security Requirements (DSR) (e.g., entity-required controls).
▪ An implemented and operational capability exists to ensure the deployment of AAT includes appropriate technical and organizational measures so that AAT are used in accordance with the AAT developer-provided instructions for use.

Level 4 Quantitatively Controlled

Utilize SCR-CMM Level 3 criteria definitions:
▪ There are no defined Level 4 criteria, since it is reasonable to assume a quantitatively-controlled process is not necessary to operationalize this control.
▪ While it may be possible to develop “metrics-driven” capabilities for this control, the criteria would be organization-specific to define.

Level 5 Continuously Improving

Utilize SCR-CMM Level 3 or Level 4 (if available) criteria definitions:
▪ There are no defined Level 5 criteria, since it is reasonable to assume a continuously-improving process is not necessary to operationalize this control.
▪ Level 5 capabilities should be considered “world-class” where the control builds on Level 4 capabilities, but are continuously improving through Artificial Intelligence (AI) and/or Machine Learning (ML) technologies.
▪ While it may be possible to develop responsive capabilities for this control through the use of AI and/or ML technologies, the criteria would be organization-specific to define.
SCF AI & Autonomous Technologies Oversight Functions Competency

Description

Mechanisms exist to ensure the deployment of Artificial Intelligence (AI) and Autonomous Technologies (AAT) assigns human oversight to individuals who have the necessary:
(1) Competence;
(2) Training;
(3) Authority; and
(4) Resources.

Possible Solutions & Considerations

Micro-Small Business (<10 staff) / BLS Firm Size Classes 1-2

∙ Train AI system operators on oversight responsibilities

Small Business (10-49 staff) / BLS Firm Size Classes 3-4

∙ Training program for personnel overseeing AI systems

Medium Business (50-249 staff) / BLS Firm Size Classes 5-6

∙ Formal AI oversight training program
∙ Competency requirements for AI oversight roles

Large Business (250-999 staff) / BLS Firm Size Classes 7-8

∙ AI training and certification program for oversight personnel

Enterprise (> 1,000 staff) / BLS Firm Size Class 9

∙ Enterprise AI training and certification program
∙ Role-based AI competency framework
∙ Continuous learning program

SCR-CMM

Level 0 Not Performed

Practices are non-existent, based on the inability to demonstrate an implemented and operational capability. A reasonable person would conclude the control is not being performed.

Level 1 Performed Informally

Artificial Intelligence and Autonomous Technology (AAT) domain capabilities are ad hoc and inconsistent. Capability criteria associated with this control may include:
▪ Policies, standards & procedures associated with AAT domain capabilities provide limited coverage due to the depth and breadth of the existing documentation.
▪ AAT-related processes are expected to follow the organization's existing processes (e.g., incident response, asset management, change control and risk assessments).
▪ No formal Governance, Risk & Compliance (GRC) team exists to provide AAT oversight, where the Chief Information Officer (CIO), or similar function, governs technology decisions what is acceptable for AAT within the organization.

Level 2 Planned Tracked

Artificial Intelligence and Autonomous Technology (AAT) capabilities are requirements-driven, but are not standardized across the entity (e.g., local/regional level consistency). Capability criteria associated with this control reasonably expect the following criteria to exist:
▪ Policies and standards associated with AAT domain capabilities are formally documented and centrally-managed by the entity.
▪ Standardized Operating Procedures (SOP) associated with AAT domain capabilities are documented and maintained by process owners.
▪ IT and/or cybersecurity personnel work with business stakeholders and process owners to appropriately scope and reasonably implement cybersecurity and data protection controls associated with AAT domain capabilities to address applicable statutory, regulatory and/or contractual requirements for Technology Assets, Applications, Services and/or Data (TAASD).
▪ Artificial Intelligence (AI)-related controls are primarily administrative and preventative in nature (e.g., policies, standards and procedures & guidelines).
▪ Asset management may be a defined function (e.g., team or department) or assigned as an additional duty to existing IT and/or cybersecurity personnel.

Level 3 Well Defined

Artificial Intelligence and Autonomous Technology (AAT) capabilities are standardized across the entity for applicability to People, Processes, Technologies, Data and/or Facilities (PPTDF) to ensure consistency for Technology Assets, Applications, Services and/or Data (TAASD). Capability criteria associated with this control reasonably expect the following criteria to exist:
▪ Policies and standards associated with AAT domain capabilities are formally documented and centrally-managed by the entity's Governance, Risk & Compliance (GRC) team, or similar function.
▪ Standardized Operating Procedures (SOP) associated with AAT domain capabilities are well-documented and kept current by process owners.
▪ An Artificial Intelligence Governance (AIG) team, or similar function, is appropriately staffed and supported to implement and maintain AAT domain capabilities.
▪ Technology is leveraged to enhance the efficiency and accuracy of AI governance, risk management and compliance operations (e.g., dedicated AI governance platform).
▪ The entity's Governance, Risk & Compliance (GRC) team, or similar function, works with business stakeholders and process owners to appropriately scope and reasonably implement cybersecurity and data protection controls associated with AAT domain capabilities to address Minimum Compliance Requirements (MCR) (e.g., applicable statutory, regulatory and/or contractual requirements) and Discretionary Security Requirements (DSR) (e.g., entity-required controls).
▪ An implemented and operational capability exists to ensure the deployment of AAT assigns human oversight to individuals who have the necessary:
(1) Competence;
(2) Training;
(3) Authority; and
(4) Resources.

Level 4 Quantitatively Controlled

Utilize SCR-CMM Level 3 criteria definitions:
▪ There are no defined Level 4 criteria, since it is reasonable to assume a quantitatively-controlled process is not necessary to operationalize this control.
▪ While it may be possible to develop “metrics-driven” capabilities for this control, the criteria would be organization-specific to define.

Level 5 Continuously Improving

Utilize SCR-CMM Level 3 or Level 4 (if available) criteria definitions:
▪ There are no defined Level 5 criteria, since it is reasonable to assume a continuously-improving process is not necessary to operationalize this control.
▪ Level 5 capabilities should be considered “world-class” where the control builds on Level 4 capabilities, but are continuously improving through Artificial Intelligence (AI) and/or Machine Learning (ML) technologies.
▪ While it may be possible to develop responsive capabilities for this control through the use of AI and/or ML technologies, the criteria would be organization-specific to define.
SCF AI & Autonomous Technologies Data Relevance

Description

Mechanisms exist to ensure the input to Artificial Intelligence (AI) and Autonomous Technologies (AAT) is relevant to the intended purpose of the AAT.

Possible Solutions & Considerations

Micro-Small Business (<10 staff) / BLS Firm Size Classes 1-2

∙ Validate that inputs to AI tools are relevant before submitting

Small Business (10-49 staff) / BLS Firm Size Classes 3-4

∙ Policy requiring relevance checking of AI inputs

Medium Business (50-249 staff) / BLS Firm Size Classes 5-6

∙ Formal AI input validation policy
∙ Input relevance criteria documentation

Large Business (250-999 staff) / BLS Firm Size Classes 7-8

∙ Input validation controls for AI systems
∙ Automated input filtering

Enterprise (> 1,000 staff) / BLS Firm Size Class 9

∙ Enterprise AI input governance framework
∙ Automated input validation and filtering
∙ Data quality controls for AI pipelines

SCR-CMM

Level 0 Not Performed

Practices are non-existent, based on the inability to demonstrate an implemented and operational capability. A reasonable person would conclude the control is not being performed.

Level 1 Performed Informally

Artificial Intelligence and Autonomous Technology (AAT) domain capabilities are ad hoc and inconsistent. Capability criteria associated with this control may include:
▪ Policies, standards & procedures associated with AAT domain capabilities provide limited coverage due to the depth and breadth of the existing documentation.
▪ AAT-related processes are expected to follow the organization's existing processes (e.g., incident response, asset management, change control and risk assessments).
▪ No formal Governance, Risk & Compliance (GRC) team exists to provide AAT oversight, where the Chief Information Officer (CIO), or similar function, governs technology decisions what is acceptable for AAT within the organization.

Level 2 Planned Tracked

Artificial Intelligence and Autonomous Technology (AAT) capabilities are requirements-driven, but are not standardized across the entity (e.g., local/regional level consistency). Capability criteria associated with this control reasonably expect the following criteria to exist:
▪ Policies and standards associated with AAT domain capabilities are formally documented and centrally-managed by the entity.
▪ Standardized Operating Procedures (SOP) associated with AAT domain capabilities are documented and maintained by process owners.
▪ IT and/or cybersecurity personnel work with business stakeholders and process owners to appropriately scope and reasonably implement cybersecurity and data protection controls associated with AAT domain capabilities to address applicable statutory, regulatory and/or contractual requirements for Technology Assets, Applications, Services and/or Data (TAASD).
▪ Artificial Intelligence (AI)-related controls are primarily administrative and preventative in nature (e.g., policies, standards and procedures & guidelines).
▪ Asset management may be a defined function (e.g., team or department) or assigned as an additional duty to existing IT and/or cybersecurity personnel.

Level 3 Well Defined

Artificial Intelligence and Autonomous Technology (AAT) capabilities are standardized across the entity for applicability to People, Processes, Technologies, Data and/or Facilities (PPTDF) to ensure consistency for Technology Assets, Applications, Services and/or Data (TAASD). Capability criteria associated with this control reasonably expect the following criteria to exist:
▪ Policies and standards associated with AAT domain capabilities are formally documented and centrally-managed by the entity's Governance, Risk & Compliance (GRC) team, or similar function.
▪ Standardized Operating Procedures (SOP) associated with AAT domain capabilities are well-documented and kept current by process owners.
▪ An Artificial Intelligence Governance (AIG) team, or similar function, is appropriately staffed and supported to implement and maintain AAT domain capabilities.
▪ Technology is leveraged to enhance the efficiency and accuracy of AI governance, risk management and compliance operations (e.g., dedicated AI governance platform).
▪ The entity's Governance, Risk & Compliance (GRC) team, or similar function, works with business stakeholders and process owners to appropriately scope and reasonably implement cybersecurity and data protection controls associated with AAT domain capabilities to address Minimum Compliance Requirements (MCR) (e.g., applicable statutory, regulatory and/or contractual requirements) and Discretionary Security Requirements (DSR) (e.g., entity-required controls).
▪ An implemented and operational capability exists to ensure the input to AAT is relevant to the intended purpose of the AAT.

Level 4 Quantitatively Controlled

Utilize SCR-CMM Level 3 criteria definitions:
▪ There are no defined Level 4 criteria, since it is reasonable to assume a quantitatively-controlled process is not necessary to operationalize this control.
▪ While it may be possible to develop “metrics-driven” capabilities for this control, the criteria would be organization-specific to define.

Level 5 Continuously Improving

Utilize SCR-CMM Level 3 or Level 4 (if available) criteria definitions:
▪ There are no defined Level 5 criteria, since it is reasonable to assume a continuously-improving process is not necessary to operationalize this control.
▪ Level 5 capabilities should be considered “world-class” where the control builds on Level 4 capabilities, but are continuously improving through Artificial Intelligence (AI) and/or Machine Learning (ML) technologies.
▪ While it may be possible to develop responsive capabilities for this control through the use of AI and/or ML technologies, the criteria would be organization-specific to define.
SCF AI & Autonomous Technologies Irregularity Reporting

Description

Mechanisms exist to ensure serious incidents and/or irregularities associated with the deployment of Artificial Intelligence (AI) and Autonomous Technologies (AAT) are reported without delay to the:
(1) AAT provider;
(2) AAT importer or distributor, if applicable; and/or
(3) Local law authorities and/or governmental agency, as required.

Possible Solutions & Considerations

Micro-Small Business (<10 staff) / BLS Firm Size Classes 1-2

∙ Report serious AI incidents and irregularities to management

Small Business (10-49 staff) / BLS Firm Size Classes 3-4

∙ Incident reporting procedure for AI irregularities

Medium Business (50-249 staff) / BLS Firm Size Classes 5-6

∙ Formal AI incident and irregularity reporting policy
∙ Regulatory reporting contacts

Large Business (250-999 staff) / BLS Firm Size Classes 7-8

∙ AI incident response program with regulatory reporting integration

Enterprise (> 1,000 staff) / BLS Firm Size Class 9

∙ Enterprise AI incident management platform
∙ Automated regulatory notification
∙ Legal and compliance integration

SCR-CMM

Level 0 Not Performed

Practices are non-existent, based on the inability to demonstrate an implemented and operational capability. A reasonable person would conclude the control is not being performed.

Level 1 Performed Informally

Artificial Intelligence and Autonomous Technology (AAT) domain capabilities are ad hoc and inconsistent. Capability criteria associated with this control may include:
▪ Policies, standards & procedures associated with AAT domain capabilities provide limited coverage due to the depth and breadth of the existing documentation.
▪ AAT-related processes are expected to follow the organization's existing processes (e.g., incident response, asset management, change control and risk assessments).
▪ No formal Governance, Risk & Compliance (GRC) team exists to provide AAT oversight, where the Chief Information Officer (CIO), or similar function, governs technology decisions what is acceptable for AAT within the organization.

Level 2 Planned Tracked

Artificial Intelligence and Autonomous Technology (AAT) capabilities are requirements-driven, but are not standardized across the entity (e.g., local/regional level consistency). Capability criteria associated with this control reasonably expect the following criteria to exist:
▪ Policies and standards associated with AAT domain capabilities are formally documented and centrally-managed by the entity.
▪ Standardized Operating Procedures (SOP) associated with AAT domain capabilities are documented and maintained by process owners.
▪ IT and/or cybersecurity personnel work with business stakeholders and process owners to appropriately scope and reasonably implement cybersecurity and data protection controls associated with AAT domain capabilities to address applicable statutory, regulatory and/or contractual requirements for Technology Assets, Applications, Services and/or Data (TAASD).
▪ Artificial Intelligence (AI)-related controls are primarily administrative and preventative in nature (e.g., policies, standards and procedures & guidelines).
▪ Asset management may be a defined function (e.g., team or department) or assigned as an additional duty to existing IT and/or cybersecurity personnel.

Level 3 Well Defined

Artificial Intelligence and Autonomous Technology (AAT) capabilities are standardized across the entity for applicability to People, Processes, Technologies, Data and/or Facilities (PPTDF) to ensure consistency for Technology Assets, Applications, Services and/or Data (TAASD). Capability criteria associated with this control reasonably expect the following criteria to exist:
▪ Policies and standards associated with AAT domain capabilities are formally documented and centrally-managed by the entity's Governance, Risk & Compliance (GRC) team, or similar function.
▪ Standardized Operating Procedures (SOP) associated with AAT domain capabilities are well-documented and kept current by process owners.
▪ An Artificial Intelligence Governance (AIG) team, or similar function, is appropriately staffed and supported to implement and maintain AAT domain capabilities.
▪ Technology is leveraged to enhance the efficiency and accuracy of AI governance, risk management and compliance operations (e.g., dedicated AI governance platform).
▪ The entity's Governance, Risk & Compliance (GRC) team, or similar function, works with business stakeholders and process owners to appropriately scope and reasonably implement cybersecurity and data protection controls associated with AAT domain capabilities to address Minimum Compliance Requirements (MCR) (e.g., applicable statutory, regulatory and/or contractual requirements) and Discretionary Security Requirements (DSR) (e.g., entity-required controls).
▪ An implemented and operational capability exists to ensure serious incidents and/or irregularities associated with the deployment of AAT are reported without delay to the:
(1) AAT provider;
(2) AAT importer or distributor, if applicable; and/or
(3) Local law authorities and/or governmental agency, as required.

Level 4 Quantitatively Controlled

Utilize SCR-CMM Level 3 criteria definitions:
▪ There are no defined Level 4 criteria, since it is reasonable to assume a quantitatively-controlled process is not necessary to operationalize this control.
▪ While it may be possible to develop “metrics-driven” capabilities for this control, the criteria would be organization-specific to define.

Level 5 Continuously Improving

Utilize SCR-CMM Level 3 or Level 4 (if available) criteria definitions:
▪ There are no defined Level 5 criteria, since it is reasonable to assume a continuously-improving process is not necessary to operationalize this control.
▪ Level 5 capabilities should be considered “world-class” where the control builds on Level 4 capabilities, but are continuously improving through Artificial Intelligence (AI) and/or Machine Learning (ML) technologies.
▪ While it may be possible to develop responsive capabilities for this control through the use of AI and/or ML technologies, the criteria would be organization-specific to define.
SCF AI & Autonomous Technologies Use Notification To Employees

Description

Mechanisms exist to ensure employees, including workers' representatives, are informed about Artificial Intelligence (AI) and Autonomous Technologies (AAT) deployments, prior to the use of the AAT in a production environment.

Possible Solutions & Considerations

Micro-Small Business (<10 staff) / BLS Firm Size Classes 1-2

∙ Inform employees of AI tools being used in their work

Small Business (10-49 staff) / BLS Firm Size Classes 3-4

∙ Policy requiring employee notification of AI tools affecting their work

Medium Business (50-249 staff) / BLS Firm Size Classes 5-6

∙ Formal employee notification policy for AI deployments
∙ Communication procedures

Large Business (250-999 staff) / BLS Firm Size Classes 7-8

∙ AI transparency program for employees
∙ Works council/union notification as required

Enterprise (> 1,000 staff) / BLS Firm Size Class 9

∙ Enterprise AI employee transparency framework
∙ Formal notification procedures
∙ Labor law compliance tracking

SCR-CMM

Level 0 Not Performed

Practices are non-existent, based on the inability to demonstrate an implemented and operational capability. A reasonable person would conclude the control is not being performed.

Level 1 Performed Informally

Artificial Intelligence and Autonomous Technology (AAT) domain capabilities are ad hoc and inconsistent. Capability criteria associated with this control may include:
▪ Policies, standards & procedures associated with AAT domain capabilities provide limited coverage due to the depth and breadth of the existing documentation.
▪ AAT-related processes are expected to follow the organization's existing processes (e.g., incident response, asset management, change control and risk assessments).
▪ No formal Governance, Risk & Compliance (GRC) team exists to provide AAT oversight, where the Chief Information Officer (CIO), or similar function, governs technology decisions what is acceptable for AAT within the organization.

Level 2 Planned Tracked

Artificial Intelligence and Autonomous Technology (AAT) capabilities are requirements-driven, but are not standardized across the entity (e.g., local/regional level consistency). Capability criteria associated with this control reasonably expect the following criteria to exist:
▪ Policies and standards associated with AAT domain capabilities are formally documented and centrally-managed by the entity.
▪ Standardized Operating Procedures (SOP) associated with AAT domain capabilities are documented and maintained by process owners.
▪ IT and/or cybersecurity personnel work with business stakeholders and process owners to appropriately scope and reasonably implement cybersecurity and data protection controls associated with AAT domain capabilities to address applicable statutory, regulatory and/or contractual requirements for Technology Assets, Applications, Services and/or Data (TAASD).
▪ Artificial Intelligence (AI)-related controls are primarily administrative and preventative in nature (e.g., policies, standards and procedures & guidelines).
▪ Asset management may be a defined function (e.g., team or department) or assigned as an additional duty to existing IT and/or cybersecurity personnel.

Level 3 Well Defined

Artificial Intelligence and Autonomous Technology (AAT) capabilities are standardized across the entity for applicability to People, Processes, Technologies, Data and/or Facilities (PPTDF) to ensure consistency for Technology Assets, Applications, Services and/or Data (TAASD). Capability criteria associated with this control reasonably expect the following criteria to exist:
▪ Policies and standards associated with AAT domain capabilities are formally documented and centrally-managed by the entity's Governance, Risk & Compliance (GRC) team, or similar function.
▪ Standardized Operating Procedures (SOP) associated with AAT domain capabilities are well-documented and kept current by process owners.
▪ An Artificial Intelligence Governance (AIG) team, or similar function, is appropriately staffed and supported to implement and maintain AAT domain capabilities.
▪ Technology is leveraged to enhance the efficiency and accuracy of AI governance, risk management and compliance operations (e.g., dedicated AI governance platform).
▪ The entity's Governance, Risk & Compliance (GRC) team, or similar function, works with business stakeholders and process owners to appropriately scope and reasonably implement cybersecurity and data protection controls associated with AAT domain capabilities to address Minimum Compliance Requirements (MCR) (e.g., applicable statutory, regulatory and/or contractual requirements) and Discretionary Security Requirements (DSR) (e.g., entity-required controls).
▪ An implemented and operational capability exists to ensure employees, including workers' representatives, are informed about AAT deployments, prior to the use of the AAT in a production environment.

Level 4 Quantitatively Controlled

Utilize SCR-CMM Level 3 criteria definitions:
▪ There are no defined Level 4 criteria, since it is reasonable to assume a quantitatively-controlled process is not necessary to operationalize this control.
▪ While it may be possible to develop “metrics-driven” capabilities for this control, the criteria would be organization-specific to define.

Level 5 Continuously Improving

Utilize SCR-CMM Level 3 or Level 4 (if available) criteria definitions:
▪ There are no defined Level 5 criteria, since it is reasonable to assume a continuously-improving process is not necessary to operationalize this control.
▪ Level 5 capabilities should be considered “world-class” where the control builds on Level 4 capabilities, but are continuously improving through Artificial Intelligence (AI) and/or Machine Learning (ML) technologies.
▪ While it may be possible to develop responsive capabilities for this control through the use of AI and/or ML technologies, the criteria would be organization-specific to define.
SCF AI & Autonomous Technologies Use Notification To Users

Description

Mechanisms exist to ensure Artificial Intelligence (AI) and Autonomous Technologies (AAT) that make decisions, or assist in making decisions, inform the people in a clear manner that they are:
(1) Utilizing an AAT solution; and
(2) Expected to validate the output for relevance and accuracy.

Possible Solutions & Considerations

Micro-Small Business (<10 staff) / BLS Firm Size Classes 1-2

∙ Inform users when AI is making or assisting decisions affecting them

Small Business (10-49 staff) / BLS Firm Size Classes 3-4

∙ User notification policy for AI-assisted decisions

Medium Business (50-249 staff) / BLS Firm Size Classes 5-6

∙ Formal user notification policy and procedures for AI decision support

Large Business (250-999 staff) / BLS Firm Size Classes 7-8

∙ AI user transparency program
∙ Automated user notifications for AI decisions

Enterprise (> 1,000 staff) / BLS Firm Size Class 9

∙ Enterprise AI user transparency framework
∙ Automated disclosure mechanisms
∙ Regulatory compliance (EU AI Act, GDPR)

SCR-CMM

Level 0 Not Performed

Practices are non-existent, based on the inability to demonstrate an implemented and operational capability. A reasonable person would conclude the control is not being performed.

Level 1 Performed Informally

Artificial Intelligence and Autonomous Technology (AAT) domain capabilities are ad hoc and inconsistent. Capability criteria associated with this control may include:
▪ Policies, standards & procedures associated with AAT domain capabilities provide limited coverage due to the depth and breadth of the existing documentation.
▪ AAT-related processes are expected to follow the organization's existing processes (e.g., incident response, asset management, change control and risk assessments).
▪ No formal Governance, Risk & Compliance (GRC) team exists to provide AAT oversight, where the Chief Information Officer (CIO), or similar function, governs technology decisions what is acceptable for AAT within the organization.

Level 2 Planned Tracked

Artificial Intelligence and Autonomous Technology (AAT) capabilities are requirements-driven, but are not standardized across the entity (e.g., local/regional level consistency). Capability criteria associated with this control reasonably expect the following criteria to exist:
▪ Policies and standards associated with AAT domain capabilities are formally documented and centrally-managed by the entity.
▪ Standardized Operating Procedures (SOP) associated with AAT domain capabilities are documented and maintained by process owners.
▪ IT and/or cybersecurity personnel work with business stakeholders and process owners to appropriately scope and reasonably implement cybersecurity and data protection controls associated with AAT domain capabilities to address applicable statutory, regulatory and/or contractual requirements for Technology Assets, Applications, Services and/or Data (TAASD).
▪ Artificial Intelligence (AI)-related controls are primarily administrative and preventative in nature (e.g., policies, standards and procedures & guidelines).
▪ Asset management may be a defined function (e.g., team or department) or assigned as an additional duty to existing IT and/or cybersecurity personnel.

Level 3 Well Defined

Artificial Intelligence and Autonomous Technology (AAT) capabilities are standardized across the entity for applicability to People, Processes, Technologies, Data and/or Facilities (PPTDF) to ensure consistency for Technology Assets, Applications, Services and/or Data (TAASD). Capability criteria associated with this control reasonably expect the following criteria to exist:
▪ Policies and standards associated with AAT domain capabilities are formally documented and centrally-managed by the entity's Governance, Risk & Compliance (GRC) team, or similar function.
▪ Standardized Operating Procedures (SOP) associated with AAT domain capabilities are well-documented and kept current by process owners.
▪ An Artificial Intelligence Governance (AIG) team, or similar function, is appropriately staffed and supported to implement and maintain AAT domain capabilities.
▪ Technology is leveraged to enhance the efficiency and accuracy of AI governance, risk management and compliance operations (e.g., dedicated AI governance platform).
▪ The entity's Governance, Risk & Compliance (GRC) team, or similar function, works with business stakeholders and process owners to appropriately scope and reasonably implement cybersecurity and data protection controls associated with AAT domain capabilities to address Minimum Compliance Requirements (MCR) (e.g., applicable statutory, regulatory and/or contractual requirements) and Discretionary Security Requirements (DSR) (e.g., entity-required controls).
▪ An implemented and operational capability exists to ensure AAT that make decisions, or assist in making decisions, inform the people in a clear manner that they are:
(1) Utilizing an AAT solution; and
(2) Expected to validate the output for relevance and accuracy.

Level 4 Quantitatively Controlled

Utilize SCR-CMM Level 3 criteria definitions:
▪ There are no defined Level 4 criteria, since it is reasonable to assume a quantitatively-controlled process is not necessary to operationalize this control.
▪ While it may be possible to develop “metrics-driven” capabilities for this control, the criteria would be organization-specific to define.

Level 5 Continuously Improving

Utilize SCR-CMM Level 3 or Level 4 (if available) criteria definitions:
▪ There are no defined Level 5 criteria, since it is reasonable to assume a continuously-improving process is not necessary to operationalize this control.
▪ Level 5 capabilities should be considered “world-class” where the control builds on Level 4 capabilities, but are continuously improving through Artificial Intelligence (AI) and/or Machine Learning (ML) technologies.
▪ While it may be possible to develop responsive capabilities for this control through the use of AI and/or ML technologies, the criteria would be organization-specific to define.
SCF AI & Autonomous Technologies Event Logging

Description

Mechanisms exist to ensure Artificial Intelligence (AI) and Autonomous Technologies (AAT) system event logging capabilities at a minimum provide:
(1) Start date, start time, end date and end time for each use;
(2) Database(s) against which input data has been checked by the system;
(3) Input data for which the search has led to a match; and
(4) Identification of individual(s) involved in the verification of the results.

Possible Solutions & Considerations

Micro-Small Business (<10 staff) / BLS Firm Size Classes 1-2

∙ Enable basic logging on AI/ML systems

Small Business (10-49 staff) / BLS Firm Size Classes 3-4

∙ Configure event logging on AI systems
∙ Define log retention periods

Medium Business (50-249 staff) / BLS Firm Size Classes 5-6

∙ Formal AI event logging policy
∙ Centralized log collection

Large Business (250-999 staff) / BLS Firm Size Classes 7-8

∙ SIEM integration for AI system logs
∙ Defined logging standards for AI platforms

Enterprise (> 1,000 staff) / BLS Firm Size Class 9

∙ Enterprise SIEM platform
∙ Standardized AI event logging schema
∙ Automated alerting for anomalous AI events

SCR-CMM

Level 0 Not Performed

Practices are non-existent, based on the inability to demonstrate an implemented and operational capability. A reasonable person would conclude the control is not being performed.

Level 1 Performed Informally

Artificial Intelligence and Autonomous Technology (AAT) domain capabilities are ad hoc and inconsistent. Capability criteria associated with this control may include:
▪ Policies, standards & procedures associated with AAT domain capabilities provide limited coverage due to the depth and breadth of the existing documentation.
▪ AAT-related processes are expected to follow the organization's existing processes (e.g., incident response, asset management, change control and risk assessments).
▪ No formal Governance, Risk & Compliance (GRC) team exists to provide AAT oversight, where the Chief Information Officer (CIO), or similar function, governs technology decisions what is acceptable for AAT within the organization.

Level 2 Planned Tracked

Artificial Intelligence and Autonomous Technology (AAT) capabilities are requirements-driven, but are not standardized across the entity (e.g., local/regional level consistency). Capability criteria associated with this control reasonably expect the following criteria to exist:
▪ Policies and standards associated with AAT domain capabilities are formally documented and centrally-managed by the entity.
▪ Standardized Operating Procedures (SOP) associated with AAT domain capabilities are documented and maintained by process owners.
▪ IT and/or cybersecurity personnel work with business stakeholders and process owners to appropriately scope and reasonably implement cybersecurity and data protection controls associated with AAT domain capabilities to address applicable statutory, regulatory and/or contractual requirements for Technology Assets, Applications, Services and/or Data (TAASD).
▪ Artificial Intelligence (AI)-related controls are primarily administrative and preventative in nature (e.g., policies, standards and procedures & guidelines).
▪ Asset management may be a defined function (e.g., team or department) or assigned as an additional duty to existing IT and/or cybersecurity personnel.

Level 3 Well Defined

Artificial Intelligence and Autonomous Technology (AAT) capabilities are standardized across the entity for applicability to People, Processes, Technologies, Data and/or Facilities (PPTDF) to ensure consistency for Technology Assets, Applications, Services and/or Data (TAASD). Capability criteria associated with this control reasonably expect the following criteria to exist:
▪ Policies and standards associated with AAT domain capabilities are formally documented and centrally-managed by the entity's Governance, Risk & Compliance (GRC) team, or similar function.
▪ Standardized Operating Procedures (SOP) associated with AAT domain capabilities are well-documented and kept current by process owners.
▪ An Artificial Intelligence Governance (AIG) team, or similar function, is appropriately staffed and supported to implement and maintain AAT domain capabilities.
▪ Technology is leveraged to enhance the efficiency and accuracy of AI governance, risk management and compliance operations (e.g., dedicated AI governance platform).
▪ The entity's Governance, Risk & Compliance (GRC) team, or similar function, works with business stakeholders and process owners to appropriately scope and reasonably implement cybersecurity and data protection controls associated with AAT domain capabilities to address Minimum Compliance Requirements (MCR) (e.g., applicable statutory, regulatory and/or contractual requirements) and Discretionary Security Requirements (DSR) (e.g., entity-required controls).
▪ An implemented and operational capability exists to ensure AAT system event logging capabilities at a minimum provide:
(1) Start date, start time, end date and end time for each use;
(2) Database(s) against which input data has been checked by the system;
(3) Input data for which the search has led to a match; and
(4) Identification of individual(s) involved in the verification of the results.

Level 4 Quantitatively Controlled

Utilize SCR-CMM Level 3 criteria definitions:
▪ There are no defined Level 4 criteria, since it is reasonable to assume a quantitatively-controlled process is not necessary to operationalize this control.
▪ While it may be possible to develop “metrics-driven” capabilities for this control, the criteria would be organization-specific to define.

Level 5 Continuously Improving

Utilize SCR-CMM Level 3 or Level 4 (if available) criteria definitions:
▪ There are no defined Level 5 criteria, since it is reasonable to assume a continuously-improving process is not necessary to operationalize this control.
▪ Level 5 capabilities should be considered “world-class” where the control builds on Level 4 capabilities, but are continuously improving through Artificial Intelligence (AI) and/or Machine Learning (ML) technologies.
▪ While it may be possible to develop responsive capabilities for this control through the use of AI and/or ML technologies, the criteria would be organization-specific to define.
SCF AI & Autonomous Technologies Registration

Description

Mechanisms exist to maintain a current registration for Artificial Intelligence (AI) and Autonomous Technologies (AAT) with the appropriate governing body, as required by statutory or regulatory requirements.

Possible Solutions & Considerations

Micro-Small Business (<10 staff) / BLS Firm Size Classes 1-2

∙ Determine whether any AI use requires registration (e.g., EU AI Act high-risk systems)

Small Business (10-49 staff) / BLS Firm Size Classes 3-4

∙ Track AI registration obligations by jurisdiction in compliance checklist

Medium Business (50-249 staff) / BLS Firm Size Classes 5-6

∙ Register high-risk AI systems with relevant authorities as required

Large Business (250-999 staff) / BLS Firm Size Classes 7-8

∙ AI registration compliance program
∙ Track registration requirements by jurisdiction

Enterprise (> 1,000 staff) / BLS Firm Size Class 9

∙ Enterprise AI regulatory compliance platform
∙ Automated registration tracking
∙ EU AI Act database registration compliance

SCR-CMM

Level 0 Not Performed

Practices are non-existent, based on the inability to demonstrate an implemented and operational capability. A reasonable person would conclude the control is not being performed.

Level 1 Performed Informally

Artificial Intelligence and Autonomous Technology (AAT) domain capabilities are ad hoc and inconsistent. Capability criteria associated with this control may include:
▪ Policies, standards & procedures associated with AAT domain capabilities provide limited coverage due to the depth and breadth of the existing documentation.
▪ AAT-related processes are expected to follow the organization's existing processes (e.g., incident response, asset management, change control and risk assessments).
▪ No formal Governance, Risk & Compliance (GRC) team exists to provide AAT oversight, where the Chief Information Officer (CIO), or similar function, governs technology decisions what is acceptable for AAT within the organization.

Level 2 Planned Tracked

Artificial Intelligence and Autonomous Technology (AAT) capabilities are requirements-driven, but are not standardized across the entity (e.g., local/regional level consistency). Capability criteria associated with this control reasonably expect the following criteria to exist:
▪ Policies and standards associated with AAT domain capabilities are formally documented and centrally-managed by the entity.
▪ Standardized Operating Procedures (SOP) associated with AAT domain capabilities are documented and maintained by process owners.
▪ IT and/or cybersecurity personnel work with business stakeholders and process owners to appropriately scope and reasonably implement cybersecurity and data protection controls associated with AAT domain capabilities to address applicable statutory, regulatory and/or contractual requirements for Technology Assets, Applications, Services and/or Data (TAASD).
▪ Artificial Intelligence (AI)-related controls are primarily administrative and preventative in nature (e.g., policies, standards and procedures & guidelines).
▪ Asset management may be a defined function (e.g., team or department) or assigned as an additional duty to existing IT and/or cybersecurity personnel.
▪ AAT is regarded as a technology and governed by the entity's existing IT governance practices.
▪ No formal Governance, Risk & Compliance (GRC) team exists to provide oversight of AAT-related activities. GRC functions are assigned to existing IT and/or cybersecurity personnel.

Level 3 Well Defined

Compliance (CPL) capabilities are standardized across the entity for applicability to People, Processes, Technologies, Data and/or Facilities (PPTDF) to ensure consistency for Technology Assets, Applications, Services and/or Data (TAASD). Capability criteria associated with this control reasonably expect the following criteria to exist:
▪ Policies and standards associated with CPL domain capabilities are formally documented and centrally-managed by the entity's Governance, Risk & Compliance (GRC) team, or similar function.
▪ Standardized Operating Procedures (SOP) associated with CPL domain capabilities are well-documented and kept current by process owners.
▪ A Governance, Risk & Compliance (GRC) team, or similar function, is appropriately staffed and supported to implement and maintain CPL domain capabilities.
▪ Technology is leveraged to enhance the efficiency and accuracy of governance, risk management and compliance operations (e.g., GRC platform).
▪ The entity's Governance, Risk & Compliance (GRC) team, or similar function, works with business stakeholders and process owners to appropriately scope and reasonably implement cybersecurity and data protection controls associated with CPL domain capabilities to address Minimum Compliance Requirements (MCR) (e.g., applicable statutory, regulatory and/or contractual requirements) and Discretionary Security Requirements (DSR) (e.g., entity-required controls).
▪ An implemented and operational capability exists to maintain a current registration for AAT with the appropriate governing body, as required by statutory or regulatory requirements.

Level 4 Quantitatively Controlled

Utilize SCR-CMM Level 3 criteria definitions:
▪ There are no defined Level 4 criteria, since it is reasonable to assume a quantitatively-controlled process is not necessary to operationalize this control.
▪ While it may be possible to develop “metrics-driven” capabilities for this control, the criteria would be organization-specific to define.

Level 5 Continuously Improving

Utilize SCR-CMM Level 3 or Level 4 (if available) criteria definitions:
▪ There are no defined Level 5 criteria, since it is reasonable to assume a continuously-improving process is not necessary to operationalize this control.
▪ Level 5 capabilities should be considered “world-class” where the control builds on Level 4 capabilities, but are continuously improving through Artificial Intelligence (AI) and/or Machine Learning (ML) technologies.
▪ While it may be possible to develop responsive capabilities for this control through the use of AI and/or ML technologies, the criteria would be organization-specific to define.
SCF Data Protection Impact Assessment (DPIA)

Description

Mechanisms exist to conduct a Data Protection Impact Assessment (DPIA) on Technology Assets, Applications and/or Services (TAAS) that store, process and/or transmit Personal Data (PD) to identify and remediate reasonably-expected risks.

Possible Solutions & Considerations

Micro-Small Business (<10 staff) / BLS Firm Size Classes 1-2

∙ Risk Management Program (RMP)
∙ Data Protection Impact Assessment (DPIA)
∙ Privacy Impact Assessment (PIA)

Small Business (10-49 staff) / BLS Firm Size Classes 3-4

∙ Risk Management Program (RMP)
∙ Data Protection Impact Assessment (DPIA)
∙ Privacy Impact Assessment (PIA)

Medium Business (50-249 staff) / BLS Firm Size Classes 5-6

∙ Risk Management Program (RMP)
∙ Data Protection Impact Assessment (DPIA)
∙ Privacy Impact Assessment (PIA)

Large Business (250-999 staff) / BLS Firm Size Classes 7-8

∙ Risk Management Program (RMP)
∙ Data Protection Impact Assessment (DPIA)
∙ Privacy Impact Assessment (PIA)

Enterprise (> 1,000 staff) / BLS Firm Size Class 9

∙ Risk Management Program (RMP)
∙ Data Protection Impact Assessment (DPIA)
∙ Privacy Impact Assessment (PIA)

SCR-CMM

Level 0 Not Performed

Practices are non-existent, based on the inability to demonstrate an implemented and operational capability. A reasonable person would conclude the control is not being performed.

Level 1 Performed Informally

SCR-CMM Level 1 criteria definitions are not available for this control:
▪ A reasonable person would conclude this control requires a structured process.
▪ At this level of maturity, the "ad hoc" nature of performing a capability informally would indicate the intent of the control is not met due to a lack of consistency and formality.

Level 2 Planned Tracked

Risk Management (RSK) capabilities are requirements-driven, but are not standardized across the entity (e.g., local/regional level consistency). Capability criteria associated with this control reasonably expect the following criteria to exist:
▪ Policies and standards associated with RSK domain capabilities are formally documented and centrally-managed by the entity.
▪ Standardized Operating Procedures (SOP) associated with RSK domain capabilities are documented and maintained by process owners.
▪ IT and/or cybersecurity personnel work with business stakeholders and process owners to appropriately scope and reasonably implement cybersecurity and data protection controls associated with RSK domain capabilities to address applicable statutory, regulatory and/or contractual requirements for Technology Assets, Applications, Services and/or Data (TAASD).
▪ Risk management-related controls are primarily administrative and preventative in nature (e.g., policies, standards and procedures & guidelines).
▪ Risk management may be a defined function (e.g., team or department) or assigned as an additional duty to existing IT and/or cybersecurity personnel.
▪ Risk management processes (e.g., risk assessments) and technologies focus on protecting High Value Assets (HVAs), including environments where sensitive/regulated data is stored, transmitted and processed.
▪ IT and/or cybersecurity personnel implement and maintain a form of Risk Management Program (RMP) that provides operational guidance on how risk is identified, assessed, remediated and reported.
▪ Data/process owners are expected to self-manage risks associated with their systems, applications, services and data, based on the organization's published policies and standards, including the identification, remediation and reporting of risks.
▪ Business process owners (BPOs) are made aware of cybersecurity and data protection risk(s).

Level 3 Well Defined

Risk Management (RSK) capabilities are standardized across the entity for applicability to People, Processes, Technologies, Data and/or Facilities (PPTDF) to ensure consistency for Technology Assets, Applications, Services and/or Data (TAASD). Capability criteria associated with this control reasonably expect the following criteria to exist:
▪ Policies and standards associated with RSK domain capabilities are formally documented and centrally-managed by the entity's Governance, Risk & Compliance (GRC) team, or similar function.
▪ Standardized Operating Procedures (SOP) associated with RSK domain capabilities are well-documented and kept current by process owners.
▪ A risk management team, or similar function, is appropriately staffed and supported to implement and maintain RSK domain capabilities.
▪ Technology is leveraged to enhance the efficiency and accuracy of risk management operations (e.g., risk management solution, GRC platform and TPRM tool).
▪ The entity's Governance, Risk & Compliance (GRC) team, or similar function, works with business stakeholders and process owners to appropriately scope and reasonably implement cybersecurity and data protection controls associated with RSK domain capabilities to address Minimum Compliance Requirements (MCR) (e.g., applicable statutory, regulatory and/or contractual requirements) and Discretionary Security Requirements (DSR) (e.g., entity-required controls).
▪ An implemented and operational capability exists to conduct a Data Protection Impact Assessment (DPIA) on Technology Assets, Applications and/or Services (TAAS) that store, process and/or transmit Personal Data (PD) to identify and remediate reasonably-expected risks.

Level 4 Quantitatively Controlled

Risk Management (RSK) capabilities, in addition to being standardized across the entity and centrally managed to ensure consistency across Technology Assets, Applications, Services and/or Data (TAASD), efforts are metrics driven to provide sufficient insight for decision makers to predict optimal performance, ensure continued operations and/or identify areas for improvement. Capability criteria associated with this control reasonably expect the following criteria to exist:
▪ Applicable SCR-CMM Level 3 (Well Defined) capabilities are implemented and operational.
▪ Metrics reporting includes quantitative analysis of Key Performance Indicators (KPIs).
▪ Metrics reporting includes quantitative analysis of Key Risk Indicators (KRIs).
▪ Scope of metrics, KPIs and KRIs covers organization-wide cybersecurity and data protection controls, including functions performed by third-parties.
▪ Organizational leadership maintains a formal process to objectively review and respond to metrics, KPIs and KRIs (e.g., monthly or quarterly review).
▪ Based on metrics analysis, process improvement recommendations are submitted for review and are handled in accordance with change control processes.
▪ Business and technical stakeholders are involved in reviewing and approving proposed changes to evolve capabilities.

Level 5 Continuously Improving

Utilize SCR-CMM Level 3 or Level 4 (if available) criteria definitions:
▪ There are no defined Level 5 criteria, since it is reasonable to assume a continuously-improving process is not necessary to operationalize this control.
▪ Level 5 capabilities should be considered “world-class” where the control builds on Level 4 capabilities, but are continuously improving through Artificial Intelligence (AI) and/or Machine Learning (ML) technologies.
▪ While it may be possible to develop responsive capabilities for this control through the use of AI and/or ML technologies, the criteria would be organization-specific to define.