Standards

What a conformance report answers: a standard's version, domains and controls, each with the evidence kinds it takes, its cadence and its crosswalks, one spec each under agentspecs/standards/. The Shield's evidence map (agent_shield.conformance) reads them and gives each control one of five statuses; the standard's version is tracked and offered, never applied silently.

Standard	Id	Version	Released	Next release	Read on	Controls	Mandatory	Retired
AIUC-1	aiuc-1	2026.4	2026-10-15	2027-01-15	2026-10-15	55	42	5
ISO/IEC 42001:2023	iso-42001	2023	2023-12-18	not scheduled	2026-10-15	70	70	0

AIUC-1 (aiuc-1)

The AI agent standard of the Artificial Intelligence Underwriting Company: six principles — data and privacy, security, safety, reliability, accountability, society — each a set of requirements with core and supplemental controls, mandatory or optional, updated each quarter and certified against a version locked for the certification's year.

• Published by AIUC (Artificial Intelligence Underwriting Company); a certification holds 12 months.
• Evidence kinds: documentation (Policies, contracts, plans and procedures a person wrote and signed.), config (Code and settings that implement a control, as they run.), logs (Audit trails and monitoring: what the system recorded as it ran.), evals (A third party's evaluation report, on the standard's cadence.), report (A report the organization publishes or records.), demonstration (A screen, a notice, a walkthrough: the control seen working.).
• Crosswalks to: EU AI Act, NIST AI RMF, ISO/IEC 42001, MITRE ATLAS, OWASP Top 10 for LLM Applications, OWASP Top 10 for Agentic Applications, OWASP AIVSS, CSA AICM, SOC 2, IBM AI Risk Atlas, Cisco AI Security & Safety Framework, Colorado AI Act, California SB 53, NYC Local Law 144.

A. Data & Privacy

Control	Title	Application	Cadence	Evidence
A001	Establish input data policy	mandatory	every 12 months	A001.1 (documentation), A001.2 (config), A001.3 (documentation)
A002	Establish output data policy	mandatory	every 12 months	A002.1 (documentation), A002.2 (config)
A003	Limit AI agent data access	mandatory	every 12 months	A003.1 (config), A003.2 (config), A003.3 (config)
A004	Protect IP & trade secrets	mandatory	every 12 months	A004.1 (documentation), A004.2 (documentation), A004.3 (config), A004.4 (config)
A005	Prevent cross-customer data exposure	mandatory	every 12 months	A005.1 (documentation), A005.2 (config), A005.3 (config)
A006	Prevent PII leakage	mandatory	every 12 months	A006.1 (config), A006.2 (config)
A007	Prevent IP violations	mandatory	every 12 months	A007.1 (documentation), A007.2 (config), A007.3 (logs)
A008	Prevent leakage of credentials and secrets	mandatory	every 12 months	A008.1 (config), A008.2 (config), A008.3 (config), A008.4 (documentation), A008.5 (config)
A009	Third-party testing for data & privacy risk	mandatory	every 3 months	A009.1 (evals)

B. Security

Control	Title	Application	Cadence	Evidence
B001	Third-party testing for security risk	mandatory	every 3 months	B001.1 (evals), B001.2 (documentation)
B002	Detect adversarial input	optional	every 3 months	B002.1 (config), B002.2 (logs), B002.3 (documentation), B002.4 (config), B002.5 (config)
B003	Manage public release of technical details	optional	every 12 months	B003.1 (documentation), B003.2 (documentation)
B004	Prevent AI endpoint scraping	mandatory	every 12 months	B004.1 (config), B004.2 (config), B004.3 (report), B004.4 (documentation)
B005	Implement real-time input filtering	optional	every 12 months	B005.1 (config), B005.2 (documentation), B005.3 (demonstration), B005.4 (logs), B005.5 (documentation)
B006	Prevent unauthorized AI agent actions	mandatory	every 12 months	B006.1 (config), B006.2 (config), B006.3 (config)
B007	Enforce user access privileges to AI systems	mandatory	every 3 months	B007.1 (config), B007.2 (documentation)
B008	Protect AI system deployment environment	mandatory	every 12 months	B008.1 (config), B008.2 (config), B008.5 (config), B008.3 (config), B008.4 (config)
B009	Limit output over-exposure	mandatory	every 12 months	B009.1 (config), B009.2 (demonstration), B009.3 (config)
B010	Promote secure patterns in generated code	mandatory	every 12 months	B010.1 (config), B010.2 (config), B010.3 (config), B010.4 (config), B010.5 (config), B010.6 (config)

C. Safety

Control	Title	Application	Cadence	Evidence
C001	Define AI risk taxonomy	mandatory	every 12 months	C001.1 (documentation), C001.2 (documentation)
C002	Conduct pre-deployment testing	mandatory	every 12 months	C002.1 (documentation), C002.2 (config), C002.3 (documentation), C002.4 (documentation)
C003	Prevent harmful outputs	mandatory	every 12 months	C003.1 (config), C003.2 (config), C003.3 (config), C003.4 (documentation)
C004	Prevent out-of-scope outputs	mandatory	every 12 months	C004.1 (config), C004.2 (logs), C004.3 (demonstration)
C005	Prevent agent-specific high risk outputs	mandatory	every 12 months	C005.1 (config), C005.2 (documentation), C005.3 (config)
C006	Prevent output vulnerabilities	mandatory	every 3 months	C006.1 (config), C006.2 (demonstration), C006.3 (config)
C007	Flag high risk outputs for human review	optional	every 12 months	C007.1 (documentation), C007.2 (config), C007.3 (documentation)
C008	Monitor AI risk categories	optional	every 12 months	C008.1 (logs), C008.2 (documentation), C008.3 (config)
C009	Enable real-time feedback and intervention	optional	every 3 months	C009.1 (demonstration), C009.2 (documentation)
C010	Third-party testing for safety risk	mandatory	every 3 months	C010.1 (evals)
C011	Third-party testing for out-of-scope outputs	retired	—	—
C012	Third-party testing for customer-defined risk	retired	—	—

D. Reliability

Control	Title	Application	Cadence	Evidence
D001	Prevent hallucinated outputs	mandatory	every 12 months	D001.1 (config), D001.2 (demonstration), D001.3 (demonstration)
D002	Third-party testing for hallucinations	retired	—	—
D003	Restrict unsafe tool calls	mandatory	every 12 months	D003.1 (config), D003.2 (config), D003.3 (config), D003.4 (config), D003.5 (documentation)
D004	Third-party testing for reliability risk	mandatory	every 3 months	D004.1 (evals)

E. Accountability

Control	Title	Application	Cadence	Evidence
E001	AI failure plan for security breaches	mandatory	every 12 months	E001.1 (documentation)
E002	AI failure plan for harmful outputs	mandatory	every 12 months	E002.1 (documentation), E002.2 (documentation)
E003	AI failure plan for hallucinations	mandatory	every 12 months	E003.1 (documentation), E003.2 (documentation)
E004	Assign accountability	mandatory	every 12 months	E004.1 (documentation), E004.2 (config)
E005	Document data storage security	mandatory	every 12 months	E005.1 (documentation)
E006	Conduct vendor and model due diligence	mandatory	every 12 months	E006.1 (documentation), E006.2 (documentation)
E007	Document system change approvals	retired	—	—
E008	Review internal processes	mandatory	every 12 months	E008.1 (documentation), E008.2 (documentation)
E009	Monitor third-party access	mandatory	every 12 months	E009.1 (config), E009.2 (config)
E010	Establish AI acceptable use policy	mandatory	every 12 months	E010.1 (documentation), E010.2 (config), E010.3 (demonstration), E010.4 (documentation)
E011	Record processing locations	mandatory	every 12 months	E011.1 (documentation), E011.2 (documentation)
E012	Document regulatory compliance	mandatory	every 6 months	E012.1 (documentation)
E013	Implement quality management system	optional	every 12 months	E013.1 (documentation), E013.2 (documentation), E013.3 (config), E013.4 (documentation), E013.5 (documentation)
E014	Share transparency reports	retired	—	—
E015	Log AI system activity	mandatory	every 12 months	E015.1 (config), E015.3 (config), E015.4 (config), E015.2 (config)
E016	Implement AI disclosure mechanisms	mandatory	every 12 months	E016.1 (demonstration)
E017	Document system transparency policy	optional	every 12 months	E017.1 (documentation), E017.2 (documentation), E017.3 (documentation)
E018	Third-party testing for accountability risk	mandatory	every 3 months	E018.1 (evals)

F. Society

Control	Title	Application	Cadence	Evidence
F001	Prevent AI cyber misuse	mandatory	every 12 months	F001.1 (documentation), F001.2 (config), F001.3 (documentation)
F002	Prevent catastrophic misuse	mandatory	every 12 months	F002.1 (documentation), F002.2 (config)

ISO/IEC 42001:2023 (iso-42001)

The international standard for an AI management system: clauses 4 to 10 of the management system and the Annex A controls. Read here through AIUC-1's crosswalk: each clause names the AIUC-1 controls that cover it and how far; what no AIUC-1 control covers is the organization's.

• Published by ISO/IEC; a certification holds 36 months.
• Evidence kinds: documentation (Policies, contracts, plans and procedures a person wrote and signed.), config (Code and settings that implement a control, as they run.), logs (Audit trails and monitoring: what the system recorded as it ran.), evals (A third party's evaluation report, on the standard's cadence.), report (A report the organization publishes or records.), demonstration (A screen, a notice, a walkthrough: the control seen working.).
• Crosswalks to: AIUC-1.
• Each control names the aiuc-1 controls that cover it, and how far.

4. Context of the organization

Clause	Title	Covered by	Coverage
4.1	Understanding the organization and its context	C001, E010	partial: AIUC-1 is focused on defining acceptable use and assessing risk of the AI system
4.2	Understanding the needs and expectations of interested parties	—	none: Outside the scope of AIUC-1
4.3	Determining the scope of the AI management system	E017, E010	full
4.4	AI management system	E013	full: Met by organizations that opt-in to the optional quality management system requirement

5. Leadership

Clause	Title	Covered by	Coverage
5.1	Leadership and commitment	E004	full
5.2	AI policy	E010, E017	full: AI Acceptable Use Policy and Transparency Policy collectively fulfil the ISO control. Organizations may develop a standalone AI policy
5.3	Roles, responsibilities and authorities	E004	full

6. Planning

Clause	Title	Covered by	Coverage
6.1.1	Actions to address risks and opportunities — General	C001, C008	full
6.1.2	AI risk assessment	C001, C008	full
6.1.3	AI risk treatment	C001, C008	full
6.1.4	AI system impact assessment	E013, C001	partial: AIUC-1 requires risk assessment and organizations can opt-in to the optional quality management system requirement, but does not require impact assessment specifically
6.2	AI objectives and planning to achieve them	—	none: Requiring high-level AI objectives is outside the scope of AIUC-1
6.3	Planning of changes	E007, E008	full: AIUC-1 requires that both system changes and process changes are reviewed

7. Support

Clause	Title	Covered by	Coverage
7.1	Resources	E013	partial: Organizations opting-in to the optional quality management system requirement will be required to documenting resource management, including security-of-supply related measures
7.2	Competence	E004	partial: AIUC-1 requires that accountability is clearly assigned, but does not require competence assessment specifically
7.3	Awareness	—	none: Internal training processes are outside the scope of AIUC-1
7.4	Communication	E014	partial: AIUC-1 does not specify internal communication requirements
7.5.1	Documented information — General	E013	full: Organizations opting-in to the optional quality management system requirement will meet this control
7.5.2	Creating and updating documented information	E008	full: The combination of quarterly internal reviews and annual recertification requirements fulfils this control
7.5.3	Control of documented information	—	full: Required as part of AIUC-1 certification process

8. Operation

Clause	Title	Covered by	Coverage
8.1	Operational planning and control	E013	full: Organizations opting-in to the optional quality management system requirement will meet this control
8.2	AI risk assessment	C001, C008	full
8.3	AI risk treatment	C001, C008	full
8.4	AI system impact assessment	E013, C001	partial: AIUC-1 requires risk assessment and organizations can opt-in to the optional quality management system requirement, but does not require impact assessment specifically

9. Performance evaluation

Clause	Title	Covered by	Coverage
9.1	Monitoring, measurement, analysis and evaluation	E013, C008	full: Organizations opting-in to the optional quality management system requirement will meet this control
9.2.1	Internal audit - General	E008	full
9.2.2	Internal audit programme	E008	full
9.3.1	Management review - General	E008	full
9.3.2	Management review inputs	E008	partial: AIUC-1 internal review requirement is smaller in scope, so fulfilment of 9.3.2 depends on organizational implementation
9.3.3	Management review results	E008	partial: AIUC-1 internal review requirement is smaller in scope, so fulfilment of 9.3.2 depends on organizational implementation

10. Improvement

Clause	Title	Covered by	Coverage
10.1	Continual improvement	E013	full: Organizations opting-in to the optional quality management system requirement will meet this control
10.2	Nonconformity and corrective action	E012, E013	full: Organizations opting-in to the optional quality management system requirement will meet this control

A. Annex A controls

Clause	Title	Covered by	Coverage
A.2.2	AI policy	E010, E017	full: AI Acceptable Use Policy and Transparency Policy collectively fulfil the ISO control. Organizations may develop a standalone AI policy
A.2.3	Alignment with other organizational policies	E012, E008	partial: AIUC-1 does not require review of internal policies specifically, but require regulatory compliance reviews and regular internal reviews
A.2.4	Review of the AI policy	E008, E017, E010	full
A.3.2	AI roles and responsibilities	E004	full
A.3.3	Reporting of concerns	E008	partial: AIUC-1 does not require internal reporting processes specifically, but require regular internal reviews
A.4.2	Resource documentation	E017, E013	partial: AIUC-1 does not require resource documentation specifically, but requires maintaining a centralized repository of system documentation and documenting resource management for organizations opting in to the optional quality management system requirement
A.4.3	Data resources	E017	full: Fulfilled by organisations opting into the capability-specific requirement A103: Track training data lineage. In addition, E017: Document system transparency policy lists documenting datasheets in control activities
A.4.4	Tooling resources	E017	full: Included in transparency policy control activities
A.4.5	System and computing resources	E017	full: Included in transparency policy control activities
A.4.6	Human resources	E004	partial: AIUC-1 is scoped to require documentation of clear accountability assigned for system changes
A.5.2	AI system impact assessment process	C001, E013	partial: AIUC-1 requires risk assessment and organizations can opt-in to the optional quality management system requirement, but does not require impact assessment specifically
A.5.3	Documentation of AI system impact assessments	C001, E013	partial: AIUC-1 requires risk assessment and organizations can opt-in to the optional quality management system requirement, but does not require impact assessment specifically
A.5.4	Assessing AI system impact on individuals or groups of individuals	C001, C008, E013	partial: AIUC-1 requires risk assessment and organizations can opt-in to the optional quality management system requirement, but does not require impact assessment specifically
A.5.5	Assessing societal impacts of AI systems	C001, F001, F002	partial: AIUC-1 requires risk assessment, but does not require impact assessment specifically. Impact on society is scoped to cyber and catastrophic misuse prevention
A.6.1.2	Objectives for responsible development of AI system	C007	none: Organizations pursuing AIUC-1 are naturally aligned with responsible AI principles, but AIUC-1 does not require documentation of responsible AI objectives specifically
A.6.1.3	Processes for responsible AI system design and development	—	none: Organizations pursuing AIUC-1 are naturally aligned with responsible AI principles, but AIUC-1 does not require documentation of responsible AI processes specifically
A.6.2.2	AI system requirements and specification	E004, E007	full
A.6.2.3	Documentation of AI system design and development	E017, E013	full
A.6.2.4	AI system verification and validation	E007, C002, C010, C011, C012, D002, D004	full
A.6.2.5	AI system deployment	C002	full
A.6.2.6	AI system operation and monitoring	C008	partial: AIUC-1 is focused on monitoring risk categories and does not have product performance monitoring/product quality metrics in scope
A.6.2.7	AI system technical documentation	E014, E013	full
A.6.2.8	AI system recording of event logs	E015	full
A.7.2	Data for development and enhancement of AI system	A001	full: Met by organizations who opt-in to the capability specific requirements on training data quality and lineage
A.7.3	Acquisition of data	A001	full: Met by organizations who opt-in to the capability specific requirement E103: Track training data lineage
A.7.4	Quality of data for AI systems	—	full: Met by organizations who opt-in to the capability specific requirement on training data quality
A.7.5	Data provenance	E011, A007	full
A.7.6	Data preparation	—	partial: AIUC-1 does not require data preparation methods specifically, but covers training data quality more broadly
A.8.2	System documentation and information for users	E014, E016	full
A.8.3	External reporting	C009	partial: AIUC-1 requirement focuses on AI system users and does not require reporting capabilities for external parties
A.8.4	Communication of incidents	E001, E003, E002	full
A.8.5	Information for interested parties	E001, E012, E014	full
A.9.2	Processes for responsible use of AI systems	E010, C007, C008	full: AIUC-1 requires an Acceptable Use policy which covers prohibited use cases to ensure responsible use of AI systems. In addition, AIUC-1 has a number of concrete requirements ensuring responsible use of the AI system including, e.g., flagging high risk recommendations for human review and monitoring AI risk categories
A.9.3	Objectives for responsible use of AI system	E010, C007	partial: Partially covered by Acceptable Use Policy, but AIUC-1 does not require documenting objectives specifically
A.9.4	Intended use of the AI system	E010, C008	full: Acceptable Use Policy control activities require implementing detection and monitoring tools to track policy violations
A.10.2	Allocating responsibilities	E004	full
A.10.3	Suppliers	E006	full
A.10.4	Customers	—	none: AIUC-1 does not have specific requirements concerning customer expectations and needs, but covers a broad range of safeguards required to protect users of AI systems including regular testing that safeguards work

The organization's controls

The controls a person writes and signs start from a template under standards/templates/, with what the catalogue knows filled in:

• E001: e001-failure-plan-security-breaches.md
• E002: e002-failure-plan-harmful-outputs.md
• E003: e003-failure-plan-hallucinations.md
• E010: e010-acceptable-use.md
• E011: e011-processing-locations.md
• E012: e012-regulatory-compliance.md
• A001: a001-input-data-policy.md
• A002: a002-output-data-policy.md
• C001: c001-risk-taxonomy.md
• E006: e006-vendor-due-diligence.md