Human-Aligned Responsible Technology — Management System
Information Technology — Artificial Intelligence — Human Impact Management System
Scope
- Applies to any organization, regardless of size, type, or nature, that provides or uses AI systems and employs human workers whose roles, tasks, or skills are affected by them.
- Recognizes two classes of applicable organization: Providers, who design or supply AI systems, and Deployers, who implement AI systems within their own workforce.
- Depth and formality of implementation shall be proportionate to the scale, complexity, and human impact profile of the organization's AI activities.
- Does not apply to AI systems used solely for internal research, sandboxed pilots, or proof-of-concept testing that are not deployed to affect live human workers.
- Intended to complement, not replace, ISO/IEC 42001, the NIST AI RMF, and jurisdiction-specific regulation such as the EU AI Act.
Table of contents
- Foreword Foreword
- Intro Introduction
- 1 Scope
- 2 Normative References
- 3 Terms and Definitions
- 4 Context of the Organization
- 5 Leadership
- 6 Planning
- 7 Support
- 8 Operation — ALIM, HEM, and EDT
- 9 Performance Evaluation
- 10 Improvement
- Annex A Reference Control Objectives and Controls (normative)
- Annex B Implementation Guidance for Human Impact Controls (informative)
- Annex C Cradle-to-Grave Lifecycle Integration (informative)
- Annex D Certification Tiers and Market Signal (informative)
- Annex E Regulatory and Framework Alignment (informative)
- Bibliography Bibliography
Intended audience
- AI product companies and AI-enhanced service providers (Providers)
- Enterprises and government contractors deploying AI into their own workforce (Deployers)
- Trust & Safety, Compliance, and AI governance leads responsible for human impact oversight
- Independent assessment and certification bodies evaluating HART conformance
- Procurement and regulatory reviewers assessing human-impact governance claims
Certification tiers
Independent assessment bodies — not Heart and Logic Foundation itself — conduct HART conformance certification. The Foundation maintains the standard and its conformance criteria.
| Tier | Requirements | Verification | Market signal |
|---|---|---|---|
| HART Foundation | Documented Human-Centered AI Policy; ALIM completed for all in-scope AI systems; EDT Threshold Register established; HEM baseline established; self-assessment against Annex A controls. | Self-assessment, reviewed by Heart and Logic Foundation | "We have committed to human-centered AI and implemented foundational governance." |
| HART Certified | All Foundation requirements met; full documentation suite; ALIM/HEM/EDT demonstrated across at least two full deployment cycles; regulatory alignment documented. | Third-party assessment by a HART-licensed assessor | "We have proven human-centered AI governance through independent verification." |
| HART Excellence | All Certified requirements met; demonstrated leadership in human impact innovation; continuous improvement across multiple cycles; advanced integration with ISO/IEC 42001. | Third-party assessment with enhanced scrutiny | "We lead the field in human-centered AI practice." |
Edition and version history
| Version | Date | Notes |
|---|---|---|
| v1.0 | July 2026 | First edition. Initial publication under Heart and Logic Foundation governance, building on the original methodology developed by Theobridge. |
The HART family of standards
HART-STD-001 is the general, cross-sector core standard. Heart and Logic Foundation is developing sector-specific companion standards that apply the same ALIM / HEM / EDT discipline to sectors with distinct human-impact considerations.
HART-Creative
Consent, attribution, and value-share for artists whose work trains or shapes generative models
Currently drafted as Annex G (Creative Attribution) of HART Core; may graduate into its own standard.
HART-Finance
Impact on both employees and the people being scored or assessed by AI-driven lending and underwriting systems
Does not cover fair-lending law itself — that remains the lane of existing banking regulation.
HART-Health
Impact on clinical staff's judgment and diagnostic skill, and deskilling risk in patient-facing roles
Does not cover clinical safety or efficacy — that remains the lane of FDA and HIPAA.
HART-Physical / Robotics
Erosion of manual and spatial judgment, and safety-critical human oversight, in physically automated environments
Does not cover OSHA or equipment safety standards themselves — those remain their own lane.
HART-Public Sector
Impact on civil servants delivering services and on the public receiving them
Does not cover administrative law or procurement rules themselves — those remain their own lane.
Get the standard
Delivered automatically by email after purchase, usually within minutes.