This standard specifies technical requirements and criteria for the evaluation of the trustworthiness of generative and agentic Artificial Intelligence (AI) systems. The standard applies to enterprise environments, in all economic, policy, and regulatory sectors. It is illustrated with Enterprise Resource Planning (ERP) and related enterprise workflows as primary examples, while remaining explicitly applicable to other enterprise AI-enabled systems, including Customer Relationship Management (CRM), Human Resources Information System (HRIS), finance, customer service, and operations platforms. The emphasis on ERP is illustrative rather than restrictive. The standard addresses: • A taxonomy that includes definitions for output integrity failures, including unsupported claims, contradictions, and source amnesia, omission of critical information, misattribution/citation errors, and agentic failure modes (e.g., inappropriate tool use, unintended task sequencing). • Measurement metrics that quantify reliability, such as unsupported-claim rates, citation coverage, and provenance completeness, grounding-attribution accuracy, reproducibility under controlled prompts, and failure-mode frequency by task class. • Technical and socio-technical (such as procedural) requirements that control retrieval grounding, prompt-to-action traceability, continuous monitoring, human-in-the-loop verification, and low-confidence escalation. • Governance artifacts specifying conformance evidence, including versioned evaluation sets, model cards, and trace logs. The standard includes requirements for continuous monitoring, time-bounded escalation logging, and escalation mechanisms, across the full life cycle, but excludes the specification of autonomous control-feedback-loop design, multi-step autonomous planning controls, or corrective agent behavior mechanisms.
- Standard Committee
- SSIT/SC - Social Implications of Technology Standards Committee
- Joint Sponsors
-
C/AISC
- Status
- Active PAR
- PAR Approval
- 2026-03-26
Working Group Details
- Society
- IEEE Society on Social Implications of Technology
- Standard Committee
- SSIT/SC - Social Implications of Technology Standards Committee
- Working Group
-
TrustGenAI - Trustworthy Enterprise Generative AI Working Group
- IEEE Program Manager
- Malia Zaman
Contact Malia Zaman - Working Group Chair
- Driss LAMRANI
Other Activities From This Working Group
Current projects that have been authorized by the IEEE SA Standards Board to develop a standard.
No Active Projects
Standards approved by the IEEE SA Standards Board that are within the 10-year lifecycle.
No Active Standards
These standards have been replaced with a revised version of the standard, or by a compilation of the original active standard and all its existing amendments, corrigenda, and errata.
No Superseded Standards
These standards have been removed from active status through a ballot where the standard is made inactive as a consensus decision of a balloting group.
No Inactive-Withdrawn Standards
These standards are removed from active status through an administrative process for standards that have not undergone a revision process within 10 years.
No Inactive-Reserved Standards