AIUC-1
Context
IntroductionCertificate overview
Framework comparisons
ChangelogAIUC-1 ConsortiumProvide input on AIUC-1Contact
Standard
A. Data & Privacy
B. Security
C. Safety
Define AI risk taxonomyConduct pre-deployment testingPrevent harmful outputsPrevent out-of-scope outputsPrevent customer-defined high risk outputsPrevent output vulnerabilitiesFlag high risk outputs for human reviewMonitor AI risk categoriesEnable real-time feedback and interventionThird-party testing for harmful outputsThird-party testing for out-of-scope outputsThird-party testing for customer-defined risk
D. Reliability
E. Accountability
F. Society
Certification
AIUC-1 certification Scoping Accredited auditors FAQ
Evidence overview
AIUC-1

Share your details and let us know how you hope to use AIUC-1

I am interested in...

The Security, Safety, and Reliability standard for AI agents

Stay up to date with AIUC-1

AIUC-1
AIUC-1.COM

© 2026.AIUC

OverviewChangelogConsortium

LEGAL

Privacy PolicyTerms of Service
AIUC-1 Standard
→
C. Safety
→
C010. Third-party testing for harmful outputs
C010

Third-party testing for harmful outputs

Appoint expert third parties to evaluate system robustness to harmful outputs including distressed outputs, angry responses, high-risk advice, offensive content, bias, and deception at least every 3 months

Keywords

Harmful OutputsDistressedAngryAdviceOffensiveBiasRisk SeverityToxigenThird-Party Testing

Application

Mandatory

Frequency

Every 3 months

Type

Preventative

Crosswalks

NIST AI RMF
GOVERN 4.3: Testing and incident sharing
MANAGE 2.2: Deployed system value
MEASURE 1.3: Independent assessment
MEASURE 2.1: TEVV documentation
MEASURE 2.6: Safety evaluation
MEASURE 2.11: Fairness and bias
MEASURE 4.1: Context-specific measurement
MEASURE 4.2: Trustworthiness validation
EU AI Act
Article 9: Risk Management System
CSA AICM
GRC-11: Bias and Fairness Assessment
A&A-02: Independent Assessments
TVM-06: Penetration Testing
ISO 42001
A.6.2.4: AI system verification and validation
IBM AI Risk Atlas
IBM 62: Output - Toxic output
IBM 58: Output - Decision bias
IBM 59: Output - Output bias
IBM 60: Output - Harmful output
IBM 84: Non-Technical - Unrepresentative risk testing
Cisco AI Security Framework
AITech-15.1: Harmful Content
NYC LL 144
5-301: Conduct Bias Audit
5-302: Manage Audit Data
CO AI Act
6-1-1703: Deployer Duties

Control activities

Typical evidence

Appointing qualified third-party assessors. Including selecting assessors with relevant technical capabilities for identified risk areas, maintaining records of assessor qualifications and independence.
Conducting regular testing. Including performing assessments of harmful outputs at least every quarter, defining testing scope and methodologies based on risk classifications and industry benchmarks like ToxiGen, coordinating with internal security and testing teams.
Maintaining documentation. Including testing scope, results, and remediation actions taken, tracking follow-up activities and resolution timelines.
C010.1 Report: Harmful output testing

Third-party evaluation report showing harmful output testing - must include documentation of assessor qualifications, testing methodology and findings, and improvement tracking with remediation timelines and documentation.

Category

Third-party Evals
Third-party evaluation report
Text-generationVoice-generationImage-generation

Organizations can submit alternative evidence demonstrating how they meet the requirement.

AIUC-1 is built with industry leaders

Phil Venables

"We need a SOC 2 for AI agents— a familiar, actionable standard for security and trust."

Google Cloud
Phil Venables
Former CISO of Google Cloud
Dr. Christina Liaghati

"Integrating MITRE ATLAS ensures AI security risk management tools are informed by the latest AI threat patterns and leverage state of the art defensive strategies."

MITRE
Dr. Christina Liaghati
MITRE ATLAS lead
Hyrum Anderson

"Today, enterprises can't reliably assess the security of their AI vendors— we need a standard to address this gap."

Cisco
Hyrum Anderson
Senior Director, Security & AI
Prof. Sanmi Koyejo

"Built on the latest advances in AI research, AIUC-1 empowers organizations to identify, assess, and mitigate AI risks with confidence."

Stanford
Prof. Sanmi Koyejo
Lead for Stanford Trustworthy AI Research
John Bautista

"AIUC-1standardizes how AI is adopted. That's powerful."

Orrick
John Bautista
Partner at Orrick
Lena Smart

"An AIUC-1certificate enables me to sign contracts much faster— it's a clear signal I can trust."

SecurityPal
Lena Smart
Head of Trust for SecurityPal and former CISO of MongoDB