The international conversation about artificial intelligence regulation is undergoing a fundamental shift in venue and power dynamics. Rather than pursuing binding treaties through the United Nations, governance decisions are increasingly being made within the International Telecommunication Union's technical standards committees, where corporate representatives significantly outnumber civil society advocates.

According to AI Weekly, this migration carries substantial implications for how nations will eventually regulate AI systems. The technical defaults and architectural choices embedded in ITU standards effectively become the baseline that national regulators inherit and build upon. This means the decisions made in these working groups today will constrain policy options available to governments months or years from now.

The Standards Setting Advantage

Standards bodies operate differently than traditional treaty negotiations. They focus on technical specifications rather than broad policy principles, and their proceedings attract less public scrutiny. Industry participation is expected and encouraged, while representation from civil society organizations, labor groups, and public interest advocates remains minimal.

  • ITU working groups include disproportionate representation from technology companies and telecom firms
  • Standards lock in technical choices that become difficult to reverse later
  • National governments typically adopt ITU standards with minimal modification
  • The process moves faster than treaty negotiation, with less opportunity for debate

The shift away from UN treaty processes reflects practical realities. Negotiating binding international agreements on AI has proven contentious, with countries unable to agree on fundamental principles. Standards work offers an alternative path that achieves coordination without requiring consensus on controversial questions like intellectual property, data sovereignty, or AI safety priorities.

A Narrow Window for Influence

The immediate challenge facing policymakers and advocacy organizations is that this transition is happening with minimal public awareness. Leadership teams must identify which ITU technical committees are shaping AI-relevant standards and ensure their voices are represented in those discussions before foundational choices are finalized.

The technical defaults set in these working groups are what national regulators will inherit, making early participation critical for shaping outcomes.

Organizations focusing exclusively on high-level UN AI governance processes risk missing where the actual decision-making has relocated. Several active ITU working groups are currently developing standards for AI system transparency, algorithmic accountability, and data handling practices. Once these technical specifications are published, governments implementing AI regulation will find themselves constrained by the architectural choices already embedded in the standards.

What Comes Next

Stakeholders concerned about ensuring AI governance remains balanced and accountable to public interest considerations need to engage immediately with ITU working group processes. This requires different expertise than UN policy advocacy; technical specialists who understand telecommunications standards and can translate between engineering and policy languages are essential.

The transition from treaties to standards represents a quiet but consequential reshaping of who influences AI's regulatory future. Unlike treaty negotiations, which demand public engagement and parliamentary ratification, standards work happens in technical committees where industry presence is normalized and civil society participation is rare. For those seeking to ensure AI governance serves broad public interests, that shift demands urgent attention and strategic intervention.