Module 6: Industry Standards and Frameworks
The IEEE 7000 series
Description
The IEEE 7000 series is a set of standards for ethical considerations in autonomous systems. Less practical than the NIST AI RMF for day-to-day implementation, but valuable as vocabulary and as a reference when you talk to stakeholders about ethics.
For AI engineers, the most relevant is IEEE 7000-2021 (Model Process for Addressing Ethical Concerns During System Design). We'll go as deep as necessary on that one and reference the others.
By the end you'll be able to:
- Know the main standards in the IEEE 7000 series
- Apply the IEEE 7000 process to build ethics into the design phase
- Distinguish when IEEE complements the NIST AI RMF
The standards in the series
| Standard | Title | Status | Relevance |
|---|---|---|---|
| IEEE 7000-2021 | Model Process for Addressing Ethical Concerns During System Design | Published | High |
| IEEE 7001-2021 | Transparency of Autonomous Systems | Published | Medium |
| IEEE 7002-2022 | Data Privacy Process | Published | Medium |
| IEEE 7003 | Algorithmic Bias Considerations | Draft | Medium |
| IEEE 7004 | Standard for Child and Student Data Governance | Draft | Niche |
| IEEE 7005 | Standard for Transparent Employer Data Governance | Draft | Niche |
| IEEE 7006 | Standard for Personal Data AI Agent Working Group | Draft | High (emerging) |
| IEEE 7007-2021 | Ontological Standard for Ethically Driven Robotics and Automation Systems | Published | Niche |
| IEEE 7008-2024 | Standard for Ethically Driven Nudging for Robotic, Intelligent and Autonomous Systems | Published | Niche |
| IEEE 7009 | Standard for Fail-Safe Design of Autonomous and Semi-Autonomous Systems | Draft | Medium |
| IEEE 7010-2020 | Recommended Practice for Assessing the Impact of Autonomous and Intelligent Systems on Human Well-Being | Published | High |
For typical AI engineering: focus on 7000, 7001, 7002, 7003 and 7010.
IEEE 7000-2021: The Model Process
The process in 4 phases
Phase 1: Establishing the context
- Identify the stakeholders
- Define the system in its context
- Identify the relevant values (cultural, organizational)
Phase 2: Value identification
- Stakeholders articulate their values
- Value conflicts are resolved explicitly
- Values are documented in the design specifications
Phase 3: Implementation
- Values are translated into technical requirements
- Design decisions trace back to values
- Trade-offs are documented
Phase 4: Monitoring
- Verify that the system genuinely upholds the declared values
- Periodic re-evaluation
- Update when the values change
Comparison with the NIST AI RMF
| Aspect | IEEE 7000 | NIST AI RMF |
|---|---|---|
| Focus | Values-driven design | Risk management |
| Phase | Primarily design-time | All phases, including operation |
| Methodology | Process-oriented | Function-oriented |
| Output | Value-design traceability | A risk register + mitigations |
| Practical implementation | Conceptual | More concrete |
When to use IEEE 7000: during the design phase, to make sure ethics are built in from the start. When to use the NIST AI RMF: throughout the lifecycle, for risk management.
Use both: they complement each other.
IEEE 7001: Transparency
Goal: define what transparency means for autonomous systems.
For AI engineers, the key part: 5 levels of transparency:
- Level 1: Disclosure that AI is involved
- Level 2: Information about the AI's purpose
- Level 3: The decision rationale
- Level 4: Operational details
- Level 5: Full data and algorithm access
For the Knowledge Assistant:
- Level 1: ✅ "This is an AI assistant"
- Level 2: ✅ The purpose is stated in onboarding
- Level 3: ✅ Citations + rationale provided
- Level 4: ⚠️ Operational details for admin users only
- Level 5: ❌ Not provided (and rarely required)
IEEE 7002: Data Privacy Process
It complements GDPR with a process-oriented approach. If you already implement GDPR rigorously, IEEE 7002 is validation, not new work.
IEEE 7003: Algorithmic Bias
It addresses bias considerations. There's a lot of overlap with M2 of this path. If you followed M2, you already implement IEEE 7003 in spirit.
The key principle of IEEE 7003: bias mitigation from the start, not as an afterthought.
IEEE 7010: Impact Assessment on Human Well-Being
Frameworks for evaluating broader societal impact:
- Mental health
- Physical safety
- Financial security
- Social relationships
- Civic engagement
For AI assistants: probably OK, but for AI with broader societal reach (social media algorithms, content moderation), it's critical.
Practical application: an ethics-by-design workflow
Combining IEEE 7000 with your process:
1. Project kick-off:
- Identify the stakeholders (users, indirectly affected, society)
- Run a workshop to identify the values (privacy, fairness, transparency, etc.)
- Document the values in the spec
2. Architecture phase:
- For each design decision, trace it to a value
- E.g.: "Multi-tenant isolation" traces to "privacy + fairness"
- E.g.: "Citations" traces to "transparency + explainability"
3. Development:
- Each feature is reviewed against the value requirements
- Trade-offs are documented (e.g., latency vs. explainability)
4. Pre-launch:
- Validate that the values are embodied
- User testing for value alignment
5. Post-launch:
- Monitor for value drift
- Periodic re-evaluation
- Iterate
Common traps
Trap 1 — IEEE as a replacement for NIST. No. IEEE is a complement. The NIST AI RMF is far more practical for day-to-day risk management.
Trap 2 — Treating standards as a marketing badge. "We follow IEEE 7000!" without really applying it. Substance > badge.
Trap 3 — IEEE as bureaucratic overhead. If you implement IEEE 7000 with the weight of enterprise compliance, you'll be frustrated. Adapt it to your scale.
Trap 4 — Ignoring emerging standards. IEEE 7006 (Personal Data AI Agent) is emerging. Watch its evolution.
Exercise
For your system:
- What level of IEEE 7001 transparency does it currently provide?
- Which values are fundamental (3-5)?
- Trace 3-5 design decisions to values
- Is an IEEE 7010 well-being assessment applicable?
See the solution
Transparency level: 3 (the decision rationale is provided via citations). Level 4 only for admin users.
Fundamental values:
- Privacy (tenant isolation, GDPR)
- Fairness (no bias against demographic groups)
- Transparency (citations, explainability)
- Accuracy (faithfulness, no hallucinations)
- User autonomy (the right to human review)
Design decisions traced:
- Multi-tenant DB filtering → Privacy
- Citations in every response → Transparency
- The bias audit toolkit → Fairness
- Human-in-the-loop for high-impact decisions → User autonomy
- The LLM grounded in retrieved docs → Accuracy
Is IEEE 7010 applicable?:
- Direct impact: minor (a productivity tool, low-risk decisions)
- Societal: medium (broader employment patterns)
- Not critical for assessment unless you scale significantly or expand into high-stakes domains
Summary
You learned:
- ✅ An overview of the IEEE 7000 series with the relevance of each standard
- ✅ IEEE 7000-2021's 4-phase process
- ✅ IEEE 7001's transparency levels
- ✅ When IEEE complements NIST
- ✅ The ethics-by-design workflow
Checkpoint: if you identify your values + trace design decisions to them, IEEE is integrated.
Next capsule
07 — ISO 42001: AI Management System. The most enterprise-oriented standard, but important to understand for when a client asks for it in an RFP.
Resources
- IEEE 7000 standards — official.
- IEEE 7000-2021 published.
- Value-Sensitive Design — the research foundation.
- IEEE Ethics in Action.