standard
AIUC-1 (Jul 2026)
AIUC-1 — AI agent security, safety and reliability standard (requirement level)
Record JSON · Open in map · Data retrieval guide
Catalog revision: 24028ffcfc2b295fa1b08ee6caa84b765f0731b321496bf4f548c49ad2177028. A connection does not establish full coverage.
Attributes
- category
- aiuc-1
- authority
- framework
Details
- authority
- framework
- version
- July 15, 2026 release (quarterly update cadence)
- publicationDate
- 2025-07 (initial); 2026-07-15 (current release)
- amendmentState
- quarterly standard updates; next release scheduled 2026-10-15
- effectiveDate
- Not provided
- source_url
- https://www.aiuc-1.com/
- reviewed_at
- 2026-09-10
- note
- AIUC certification standard for deployed AI agents and systems. The July 15, 2026 release has 51 live requirements: 43 mandatory and 8 optional. E007 and E014 are retained as withdrawn placeholders. Numbered sub-controls (for example A008.1) are not modeled. Requirement status, frequency and source links were verified against the official pages on 2026-09-10. The community navigator still uses its April 2026 snapshot: its 51 rows include E007 and E014 and omit A008 and B010. Applicability depends on the agent capabilities and audit scope.
- propositions
Source
Connections
- C005 — Prevent agent-specific high risk outputs belongs_to AIUC-1 (Jul 2026)
- C008 — Monitor AI risk categories belongs_to AIUC-1 (Jul 2026)
- A006 — Prevent PII leakage belongs_to AIUC-1 (Jul 2026)
- A002 — Establish output data policy belongs_to AIUC-1 (Jul 2026)
- A008 — Prevent leakage of credentials and secrets belongs_to AIUC-1 (Jul 2026)
- A005 — Prevent cross-customer data exposure belongs_to AIUC-1 (Jul 2026)
- D003 — Restrict unsafe tool calls belongs_to AIUC-1 (Jul 2026)
- E015 — Log AI system activity belongs_to AIUC-1 (Jul 2026)
- C001 — Define AI risk taxonomy belongs_to AIUC-1 (Jul 2026)
- E004 — Assign accountability belongs_to AIUC-1 (Jul 2026)
- C004 — Prevent out-of-scope outputs belongs_to AIUC-1 (Jul 2026)
- B009 — Limit output over-exposure belongs_to AIUC-1 (Jul 2026)
- C009 — Enable real-time feedback and intervention belongs_to AIUC-1 (Jul 2026)
- E010 — Establish AI acceptable use policy belongs_to AIUC-1 (Jul 2026)
- C012 — Third-party testing for customer-defined risk belongs_to AIUC-1 (Jul 2026)
- A003 — Limit AI agent data access belongs_to AIUC-1 (Jul 2026)
- C002 — Conduct pre-deployment testing belongs_to AIUC-1 (Jul 2026)
- B001 — Third-party testing of adversarial robustness belongs_to AIUC-1 (Jul 2026)
- D001 — Prevent hallucinated outputs belongs_to AIUC-1 (Jul 2026)
- A004 — Protect IP & trade secrets belongs_to AIUC-1 (Jul 2026)
- E011 — Record processing locations belongs_to AIUC-1 (Jul 2026)
- D004 — Third-party testing of tool calls belongs_to AIUC-1 (Jul 2026)
- C010 — Third-party testing for harmful outputs belongs_to AIUC-1 (Jul 2026)
- E005 — Document data storage security belongs_to AIUC-1 (Jul 2026)
- B006 — Prevent unauthorized AI agent actions belongs_to AIUC-1 (Jul 2026)
- E016 — Implement AI disclosure mechanisms belongs_to AIUC-1 (Jul 2026)
- B007 — Enforce user access privileges to AI systems belongs_to AIUC-1 (Jul 2026)
- F001 — Prevent AI cyber misuse belongs_to AIUC-1 (Jul 2026)
- B008 — Protect AI system deployment environment belongs_to AIUC-1 (Jul 2026)
- E002 — AI failure plan for harmful outputs belongs_to AIUC-1 (Jul 2026)
- A001 — Establish input data policy belongs_to AIUC-1 (Jul 2026)
- E003 — AI failure plan for hallucinations belongs_to AIUC-1 (Jul 2026)
- E012 — Document regulatory compliance belongs_to AIUC-1 (Jul 2026)
- F002 — Prevent catastrophic misuse belongs_to AIUC-1 (Jul 2026)
- B005 — Implement real-time input filtering belongs_to AIUC-1 (Jul 2026)
- E006 — Conduct vendor due diligence belongs_to AIUC-1 (Jul 2026)
- E001 — AI failure plan for security breaches belongs_to AIUC-1 (Jul 2026)
- D002 — Third-party testing for hallucinations belongs_to AIUC-1 (Jul 2026)
- C003 — Prevent harmful outputs belongs_to AIUC-1 (Jul 2026)
- E009 — Monitor third-party access belongs_to AIUC-1 (Jul 2026)
- B002 — Detect adversarial input belongs_to AIUC-1 (Jul 2026)
- C011 — Third-party testing for out-of-scope outputs belongs_to AIUC-1 (Jul 2026)
- B004 — Prevent AI endpoint scraping belongs_to AIUC-1 (Jul 2026)
- E008 — Review internal processes belongs_to AIUC-1 (Jul 2026)
- E013 — Implement quality management system belongs_to AIUC-1 (Jul 2026)
- C007 — Flag high risk outputs for human review belongs_to AIUC-1 (Jul 2026)
- B003 — Manage public release of technical details belongs_to AIUC-1 (Jul 2026)
- C006 — Prevent output vulnerabilities belongs_to AIUC-1 (Jul 2026)
- E017 — Document system transparency policy belongs_to AIUC-1 (Jul 2026)
- B010 — Promote secure patterns in generated code belongs_to AIUC-1 (Jul 2026)
- A007 — Prevent IP violations belongs_to AIUC-1 (Jul 2026)