Welcome

Welcome to Chapter 5.2 of the CAP certification program. This chapter on Standards Development & Participation is part of Lesson 5: Policy & Standards Influence in the Level 5 (AI Leader) track.

Technical standards are the operational infrastructure of AI governance. While laws and regulations establish the goals and principles of AI oversight, technical standards define what those principles mean in practice: how to evaluate model performance, how to document AI systems for risk assessment, how to implement fairness requirements, and what transparency disclosures are sufficient. Organizations that participate in developing these standards shape the practical meaning of AI governance in ways that legislation alone cannot achieve.

This chapter is designed for AI leaders who want to engage meaningfully in the standards development landscape: understanding which bodies matter, how standards processes work, where influence is exercised, and how to build the internal capacity to participate effectively. It complements the policy advocacy chapter by addressing the technical layer of governance where much of the most consequential detailed work happens.

The AI Standards Landscape

The AI standards landscape is more complex than most AI leaders realize. Multiple bodies at different levels, international, regional, national, and sector-specific, are developing AI standards simultaneously, and the relationships among these bodies are not always well-coordinated.

International standards bodies

ISO and IEC (operating jointly as ISO/IEC) are the primary international standards development organizations for AI. The ISO/IEC JTC 1/SC 42 committee is specifically focused on AI and has produced foundational standards including ISO/IEC 22989 (AI concepts and terminology), ISO/IEC 23053 (machine learning framework), ISO/IEC 24027 (bias in AI), and ISO/IEC 42001 (AI management systems). These international standards have significant influence because national standards bodies (ANSI in the US, BSI in the UK, DIN in Germany) typically adopt them with minimal modification, and because international regulatory bodies reference them in guidance.

National bodies with international influence

NIST (National Institute of Standards and Technology) in the United States has developed the AI Risk Management Framework (AI RMF) and accompanying Playbook, which have become reference documents for both US government AI procurement and private sector governance programs globally. The AI RMF's GOVERN, MAP, MEASURE, MANAGE functions are increasingly cited by regulators worldwide as a governance baseline. IEEE has developed AI ethics standards (IEEE 7000 series) that address algorithmic transparency, personal data AI, and fail-safe design, with participation from practitioners in over 100 countries.

European standards bodies and their regulatory connection

The EU AI Act specifically mandates harmonized standards that will define what compliance looks like for high-risk AI systems. CEN-CENELEC's JTC 21 (the European AI standards committee) is developing these standards in coordination with ETSI. Organizations that participate in CEN-CENELEC JTC 21 directly influence what EU AI Act compliance requires in operational terms. This is among the highest-leverage standards participation opportunities currently available.

Sector-specific standards

Beyond the general AI standards landscape, most sectors have specialized standards activities. In healthcare, IEC 62304 (medical device software lifecycle) and FDA guidance on AI/ML-based medical devices set sector-specific requirements. In financial services, the Basel Committee and IOSCO are developing AI governance guidance for banks and investment firms. In automotive, ISO/SAE 21434 and UN/ECE WP.29 regulations cover AI in vehicles. Sector standards often have more immediate operational relevance than general AI standards and provide more focused opportunities for practitioner expertise to have direct impact.

How Standards Are Developed and Where Influence Is Exercised

Understanding the mechanics of standards development helps you identify where participation creates the most value.

The standards development process

ISO standards development follows a defined multi-stage process: a New Work Item Proposal (NWIP) is submitted and balloted by member bodies; if approved, a Working Group (WG) develops a Working Draft (WD); successive drafts are reviewed and commented on by national delegations; a Committee Draft (CD) and then a Draft International Standard (DIS) are circulated for broader comment and ballot; and a final International Standard (IS) is published after final ballot. The process typically takes 3-5 years for complex standards. Key influence points are: the NWIP stage (determining what problems the standard addresses), the early Working Draft stages (establishing the conceptual framework), and the CD comment stage (refining specific requirements).

The delegation model and how to participate

Most international standards processes work through national delegations: each country sends a delegation that represents the national position, which is typically developed through a national mirror committee. In the US, ANSI coordinates participation in ISO processes; specific technical committees are administered by organizations like INCITS (IT standards) or SES (scientific and engineering standards). To participate in ISO AI standards, organizations typically join through their national body's mirror committee, for ISO/IEC JTC 1/SC 42 in the US, this means joining the INCITS CS1 committee. National mirror committee participation is significantly more accessible than direct ISO participation and provides full influence over the national position.

The role of expert contributors vs. observers

In working group settings, influence flows disproportionately to expert contributors. Those who draft text, propose new approaches, and bring technical evidence to working group discussions. Observers who attend meetings but contribute little have minimal influence over outcomes. Organizations seeking to shape specific standards need to invest in developing internal expertise on the specific topics the standard addresses and designating staff who will actively contribute drafts and technical analysis rather than simply monitoring the process.

Fast-track adoption and the ISO/NIST relationship

In some cases, standards developed by national bodies (particularly NIST in the US) can be adopted as international standards through a fast-track process that bypasses the full ISO development cycle. This means that influencing NIST frameworks and guidance documents, a more accessible activity than ISO participation, can translate directly into international standards influence. The NIST AI RMF's development process included extensive public comment periods and workshops that provided genuine opportunities for practitioner input to shape the final framework.

Building Your Organization's Standards Participation Strategy

Standards participation requires organizational strategy, sustained resource commitment, and a clear theory of how participation creates value for your organization and for broader AI governance.

Establishing your participation priorities

Begin by mapping the standards landscape relevant to your AI activities against a priority matrix. For each relevant standards activity, assess: (1) how directly does this standard affect your compliance obligations or competitive position? (2) how significant is your organization's technical expertise in this area? (3) how early in the development process is the standard? (4) what is the quality of the current draft, is there significant room for improvement? The highest-priority standards activities are those where the standard directly affects you, you have relevant expertise, the standard is still in early drafts, and the current direction is problematic or suboptimal. Standards that are nearly final, outside your technical expertise, or only tangentially relevant to your operations deserve monitoring but not active participation.

Resource commitment and expectation-setting

Meaningful standards participation requires consistent investment over multi-year timelines. ISO working groups typically meet 2-4 times per year, with significant between-meeting work on draft documents. Budget for travel (many meetings remain in-person or hybrid), staff time (a technically engaged participant typically devotes 15-30% of working time during active phases), and coordination costs. Be realistic with leadership: standards participation investment typically produces returns measured in years, not quarters. The value is primarily strategic, regulatory shaping, competitive intelligence, and relationship building, not near-term financial return.

Developing your technical positions

Before engaging in any standards process, develop clear, well-reasoned technical positions on the questions the standard addresses. This means: understanding the current draft and its technical implications, identifying the provisions that you believe are well-designed and should be maintained, identifying provisions that are problematic and why, and developing specific alternative language that addresses the legitimate goals of the provision more effectively. Positions that arrive in working groups as vague concerns without proposed alternatives have little influence. Positions that arrive with specific proposed text and clear technical rationale are far more likely to be adopted.

Engaging civil society and underrepresented stakeholders

AI standards processes have historically been dominated by large technology companies and government bodies, with limited participation from civil society organizations, small businesses, and the communities most affected by AI systems. Organizations that make genuine efforts to bring in these perspectives, by supporting civil society participation, contributing to access funds that enable participation by organizations without travel budgets, and actively soliciting input from affected communities, both improve the quality of standards and build the credibility that comes from being seen as honest participants in inclusive processes. Purely industry-dominated AI standards invite the backlash of being labeled as self-serving, which reduces their regulatory and public acceptance.

Key Takeaways

Standards development participation is among the highest-leverage and most underutilized forms of AI governance engagement available to AI leaders. The key principles:

Technical standards define what AI governance means in practice. Laws and regulations set goals; standards define operational meaning. Organizations that shape standards shape what compliance actually requires, before regulatory enforcement creates compliance pressure.

The most valuable participation is early and substantive. Early drafts are most open to influence; expert contributors who bring proposed text have far more impact than observers. Identify priority standards activities early and invest in being genuine contributors, not just monitors.

National mirror committee participation is the entry point. Direct ISO participation is complex; national mirror committees (INCITS CS1 in the US, BSI in the UK, etc.) are accessible and provide full influence over national positions. This is the practical starting point for most organizations.

NIST engagement provides international leverage. NIST frameworks and guidance documents have significant international influence and are developed through relatively accessible public processes. Investing in NIST engagement often produces more near-term influence than comparable investment in ISO processes.

Standards participation requires multi-year commitment. There are no quick wins in standards. Budget realistically for the long-term investment that genuine influence requires, and set appropriate expectations with leadership about the timeline for returns.

What Comes Next

In the next chapter, we will cover Policy Advocacy & Government Relations, the complementary skills for engaging with legislative and regulatory processes. Standards and policy advocacy work together: standards define the technical meaning of regulatory requirements, while policy advocacy shapes the regulatory framework within which standards operate. Organizations that are effective in both domains have the most comprehensive capacity to shape the AI governance environment.

As you complete this chapter, identify one specific standards activity relevant to your organization that you have not previously engaged with. Research the current stage of development, the participation pathway, and the key technical questions being addressed. This concrete investigation will ground the more strategic policy advocacy concepts in the next chapter.