hello@tonishatagoe.com Abu Dhabi · London · Accra · New York
AI Governance

A Sovereignty Perspective on Global AI Governance

How can nations with limited domestic AI capability exercise meaningful influence over global AI governance frameworks? The Sovereignty Caucus model offers a practical path to collective negotiating power for sovereignty-committed nations.

A Sovereignty Perspective on Global AI Governance

THE SCENARIO

Five nations from the Global South convened an informal coordination group to develop common positions for multilateral AI governance negotiations. The group shared a recognition that their individual negotiating positions were insufficient to influence the frameworks being drafted by dominant AI-producing nations. Each nation faced the same structural disadvantage: frameworks drafted by those with the greatest AI capability inevitably reflected their interests, their standards, and their assumptions about what governance should look like.

For four months, the six nations coordinated their positions. They developed a shared analytical framework that identified sovereignty risks in proposed governance mechanisms. They appointed a rotating spokesperson who presented unified positions at negotiation sessions. They documented their findings in a public register that connected governance proposals to their sovereignty implications for less-capable nations. The result was measurable: three provisions in the final governance framework were amended to include safeguards that would not have existed without the coordination.

The group was not a formal institution. It was not funded by an international organisation. It operated through voluntary cooperation, shared analysis, and coordinated negotiation strategy. It demonstrated a principle that multilateral governance literature has underemphasised: sovereignty interests advance not merely through individual capacity but through collective strategic positioning.

The Question

How can nations with limited domestic AI capability exercise meaningful influence over global AI governance frameworks that are designed by those with substantially greater capability?

This article proposes a practical framework for multilateral sovereignty engagement: the Sovereignty Caucus model. An informal coalition of nations that share sovereignty commitments can generate governance influence that individual nations cannot achieve alone — not through formal institutional power, but through coordinated analysis, shared strategy, and collective negotiating discipline.

Part One: The Structural Imbalance in Global AI Governance

Global AI governance is currently structured as a multilateral forum problem: a set of institutions within which nations negotiate common standards, guidelines, and frameworks. These forums are not neutral. They are arenas where power is exercised, positions are shaped, and outcomes reflect the relative influence of participants.

The structural imbalance is straightforward: the nations with greatest AI capability are also the nations with greatest influence in governance forums. They draft proposals, chair working groups, control secretariat functions, and shape the intellectual frameworks through which governance problems are defined. Nations with limited AI capability participate in these forums, but their participation is typically reactive: they respond to agendas set by others, comment on proposals drafted by others, and negotiate within parameters established by others.

The danger is not that this arrangement is unfair — though it is — but that it produces governance outcomes systematically biased toward the interests of AI-producing nations. Nations with limited capability lack the technical expertise to fully evaluate complex governance proposals. They must rely on external analysis provided by the same nations that produced the proposals. This creates a structural governance deficit: the regulated are dependent on the regulators for understanding the regulations.

The TEE Method recognises this structural reality and proposes a response that works within the existing multilateral architecture while building toward greater sovereignty. The response is not withdrawal from multilateral engagement — it is more effective multilateral engagement through collective analysis and coordinated positioning.

Part Two: The Sovereignty Caucus Model

The Sovereignty Caucus is an informal coalition of nations sharing a commitment to sovereign AI governance. It operates on three interlocking principles: coordinated analysis, strategic coordination, and knowledge transfer.

1. Coordinated Analysis

The Caucus develops shared analytical capacity. Member nations pool expertise and commission joint assessments of governance proposals that identify sovereignty implications. This analytical work produces structured findings that any member nation can independently validate. The analytical product is sovereignty risk identification: where do proposed frameworks create dependency risks? Where do they favour AI-producing nations’ interests? Where do they foreclose options for nations at early capability stages?

2. Strategic Coordination

Member nations coordinate their negotiation positions. Rather than each nation arriving at multilateral forums with its own analysis and priorities, the Caucus develops a shared position paper identifying common interests, collective demands, and non-negotiable red lines. This coordination transforms fragmented individual voices into a coherent collective position with greater weight in negotiation dynamics.

3. Knowledge Transfer

The Caucus builds domestic capability through structured knowledge transfer. Advanced members support capacity building in less advanced members through training programmes, analytical tool sharing, and governance template development. This is not charitable assistance — it is strategic investment. A Caucus whose legitimacy depends on representative membership must elevate the capability of all members over time.

The following Sovereignty Test Matrix for AI governance provides a structured assessment tool for nations evaluating their engagement with global governance frameworks:

DomainScore (1-5)What It AssessesKey Indicators
1. Analytical Capacity___Domestic capability to evaluate AI governance proposalsStaff with AI governance expertise; local research institutions
2. Negotiating Leverage___Ability to influence framework outcomesCoalition participation; technical contributions; economic weight
3. Standards Alignment___Control over domestic standard-setting processesNational standards bodies; regulatory autonomy; implementation capacity
4. Research Capacity___Domestic AI research and development capabilityResearch institutions; talent pipeline; R&D investment; publication output
5. Coalition Positioning___Position within other sovereign governance coalitionsCaucus membership; shared positions; collective negotiating strategy

Scoring: Total score indicates multilateral governance position. Below ten: significantly disadvantaged. Eleven to fifteen: moderate positioning with clear improvement priorities. Sixteen to twenty: strong sovereignty governance position. Above twenty: leadership position in multilateral AI governance.

Red Flag Checklist

If three or more of the following apply, the nation requires urgent sovereignty governance intervention:

☐ The nation has not independently assessed the sovereignty implications of any multilateral AI governance framework it endorsed.

☐ The nation’s negotiating position mirrors the position of a dominant AI-producing nation with no independent analysis.

☐ The nation has no domestic AI research capacity and no plan to develop it.

☐ Representatives at international AI governance bodies have no technical or governance expertise.

☐ The nation has not developed a national AI sovereignty strategy to inform its multilateral positions.

☐ The nation accepts frameworks that restrict its ability to develop domestic AI capability.

☐ The nation has never coordinated its negotiating position with other nations facing similar sovereignty constraints.

☐ Delegates to AI governance forums are funded or influenced by external parties with commercial interests in the outcomes.

☐ The nation has no mechanism for parliamentary or public oversight of its AI governance commitments.

☐ National position papers on AI governance are drafted by international organisations rather than domestic institutions.

Multilateral forums are not neutral spaces. They are arenas where power is exercised, alliances are formed, and outcomes are shaped by the relative influence of participants. recognising this reality is not cynicism — it is the prerequisite for effective engagement.

Part Three: Implementing the Caucus Model

The transition from individual multilateral participation to coordinated Caucus engagement requires deliberate institutional design. The following framework provides a practical implementation guide across three phases.

Phase 1: Foundation Establishment (Months 1-3)

• Identify member nations: The initial Caucus should include three to six nations with aligned sovereignty interests and complementary analytical capabilities. Geographic diversity strengthens legitimacy without compromising coordination.

• Establish the analytical unit: Appoint or designate a joint analytical team with expertise in AI governance, technology assessment, and diplomatic negotiation. The team draws on academic, governmental, and civil society expertise.

• Develop the analytical framework: Adapt the TEE Method to the multilateral context. Create structured assessment tools that evaluate governance proposals against sovereignty criteria.

• Produce the first joint assessment: Commission an evaluation of the current multilateral AI governance landscape through a sovereignty lens.

Phase 2: Coordination Activation (Months 4-6)

• Develop a shared position paper that articulates common interests, identifies priority demands, and establishes non-negotiable red lines.

• Conduct negotiation simulations before actual multilateral sessions. Rehearse positions, prepare responses to anticipated objections, and develop contingency approaches.

• Activate the Caucus at multilateral forums by presenting coordinated positions rather than individual national submissions.

• Document and publish findings in a public register connecting governance outcomes to their sovereignty implications for accountability and civil society oversight.

Phase 3: Capability Development (Months 7-12)

• Launch knowledge transfer programmes: Advanced members provide structured training to less advanced members in AI governance analysis, negotiation strategy, and domestic implementation.

• Develop shared governance templates: National AI governance assessment templates, contract review frameworks, and procurement guidelines that member nations can adopt and adapt.

• Establish quarterly review cycles assessing progress against objectives and evaluating coordination effectiveness.

• Expand membership by identifying additional nations that could strengthen the Caucus through their participation and capability development.

The measure of multilateral success is not whether a nation achieves everything it wants. It is whether the framework that emerges serves the interests of those who have historically had the least influence over its design.

Part Four: From Analysis to Action

Analysis without implementation produces reports. Implementation without analysis produces confusion. The Caucus model integrates both: analytical rigour informs strategic action, and strategic action validates and refines analytical approach.

Week One: National Diagnosis

Conduct a Sovereignty Test Matrix assessment of your nation’s AI governance multilateral engagement. Score each domain, identify critical gaps, and document strengths to share with Caucus partners.

Week Two: Coalition Mapping

Identify three to six potential Caucus member nations with aligned sovereignty interests. Draft informal invitation for exploratory discussions about shared analytical and negotiating capacity.

Week Three: Shared Analysis

Commission an initial joint analysis of the current multilateral AI governance landscape. Identify where sovereignty risks are concentrated and where collective action could produce measurable governance improvements.

Week Four: Position Development

Develop a shared position paper articulating common interests, priority demands, and sovereignty requirements. The paper must be technically grounded and defensible in negotiation contexts.

Week Five: Negotiation Preparation

Conduct simulation exercises for upcoming multilateral sessions. Develop coordinated talking points, responses to anticipated objections, and contingency positions.

Week Six: Forum Engagement

Activate the Caucus at the multilateral forum. Present coordinated positions. Document framework provisions that advance or undermine sovereignty interests. Build the public accountability register.

Week Seven: Outcome Evaluation

Assess governance outcomes against the Caucus position paper. Document where sovereignty interests were served, where compromised, and where ignored. This evaluation informs future analytical methodology.

Week Eight: Capability Expansion

Launch the knowledge transfer programme between Caucus members. Begin developing shared governance templates and domestic implementation frameworks. Identify additional nations for membership.

A Caucus of nations with shared sovereignty commitments carries more influence in multilateral forums than the same nations acting individually — not because of the size of their economies, but because of the coherence of their analysis and the discipline of their strategy.

The Question Revisited

How can nations with limited domestic AI capability exercise meaningful influence over global AI governance frameworks?

The answer is strategic coordination. Individual nations cannot match the influence of AI-producing nations in multilateral governance forums. But coordinated coalitions can. Not through institutional power — through analytical coherence. Not through economic weight — through strategic discipline. Not through formal authority — through the collective capacity to identify, articulate, and defend sovereignty interests that individual delegations cannot represent alone.

The Sovereignty Caucus model is an immediately actionable governance strategy that nations can implement without waiting for consensus in existing multilateral bodies. Sovereignty interests advance when nations stop negotiating from individual weakness and start coordinating from collective analysis.

The question every nation should ask is not: “What can AI-producing nations offer us in governance negotiations?” but: “What can we, as sovereignty-committed nations, generate together that no individual delegation could produce alone?”

The answer to that question — and the discipline to pursue it — is the foundation of multilateral AI governance that does not merely reflect the interests of the powerful, but serves the interests of the sovereign.

This article draws on the TEE Method framework from SOVEREIGN: Who Owns the Future? by Tonisha Tagoe (Chapter 18: Multilateral Alignment, Chapter 16: TEE in National Governance).

The TEE Method provides a structured process for evaluating dependency across all five domains. Each domain assessment produces a score from one to five, generating a composite sovereignty profile that guides remediation prioritisation. Domains scoring below three require immediate intervention. Domains scoring three to four require medium-term capability building. Domains scoring five represent strengths that can be leveraged to support capability development in weaker domains. The composite scoring mechanism ensures that remediation resources are directed to the areas of greatest need while recognising that sovereignty improvement is an ecosystem-wide process, not a series of isolated fixes.

The TEE Method provides a structured process for evaluating dependency across all five domains. Each domain assessment produces a score from one to five, generating a composite sovereignty profile that guides remediation prioritisation. Domains scoring below three require immediate intervention. Domains scoring three to four require medium-term capability building. Domains scoring five represent strengths that can be leveraged to support capability development in weaker domains. The composite scoring mechanism ensures that remediation resources are directed to the areas of greatest need while recognising that sovereignty improvement is an ecosystem-wide process, not a series of isolated fixes.

The TEE Method provides a structured process for evaluating dependency across all five domains. Each domain assessment produces a score from one to five, generating a composite sovereignty profile that guides remediation prioritisation. Domains scoring below three require immediate intervention. Domains scoring three to four require medium-term capability building. Domains scoring five represent strengths that can be leveraged to support capability development in weaker domains. The composite scoring mechanism ensures that remediation resources are directed to the areas of greatest need while recognising that sovereignty improvement is an ecosystem-wide process, not a series of isolated fixes.

The TEE Method provides a structured process for evaluating dependency across all five domains. Each domain assessment produces a score from one to five, generating a composite sovereignty profile that guides remediation prioritisation. Domains scoring below three require immediate intervention. Domains scoring three to four require medium-term capability building. Domains scoring five represent strengths that can be leveraged to support capability development in weaker domains. The composite scoring mechanism ensures that remediation resources are directed to the areas of greatest need while recognising that sovereignty improvement is an ecosystem-wide process, not a series of isolated fixes.

The TEE Method provides a structured process for evaluating dependency across all five domains. Each domain assessment produces a score from one to five, generating a composite sovereignty profile that guides remediation prioritisation. Domains scoring below three require immediate intervention. Domains scoring three to four require medium-term capability building. Domains scoring five represent strengths that can be leveraged to support capability development in weaker domains. The composite scoring mechanism ensures that remediation resources are directed to the areas of greatest need while recognising that sovereignty improvement is an ecosystem-wide process, not a series of isolated fixes.

The TEE Method provides a structured process for evaluating dependency across all five domains. Each domain assessment produces a score from one to five, generating a composite sovereignty profile that guides remediation prioritisation. Domains scoring below three require immediate intervention. Domains scoring three to four require medium-term capability building. Domains scoring five represent strengths that can be leveraged to support capability development in weaker domains. The composite scoring mechanism ensures that remediation resources are directed to the areas of greatest need while recognising that sovereignty improvement is an ecosystem-wide process, not a series of isolated fixes.

The TEE Method provides a structured process for evaluating dependency across all five domains. Each domain assessment produces a score from one to five, generating a composite sovereignty profile that guides remediation prioritisation. Domains scoring below three require immediate intervention. Domains scoring three to four require medium-term capability building. Domains scoring five represent strengths that can be leveraged to support capability development in weaker domains. The composite scoring mechanism ensures that remediation resources are directed to the areas of greatest need while recognising that sovereignty improvement is an ecosystem-wide process, not a series of isolated fixes.

The TEE Method provides a structured process for evaluating dependency across all five domains. Each domain assessment produces a score from one to five, generating a composite sovereignty profile that guides remediation prioritisation. Domains scoring below three require immediate intervention. Domains scoring three to four require medium-term capability building. Domains scoring five represent strengths that can be leveraged to support capability development in weaker domains. The composite scoring mechanism ensures that remediation resources are directed to the areas of greatest need while recognising that sovereignty improvement is an ecosystem-wide process, not a series of isolated fixes.

The TEE Method provides a structured process for evaluating dependency across all five domains. Each domain assessment produces a score from one to five, generating a composite sovereignty profile that guides remediation prioritisation. Domains scoring below three require immediate intervention. Domains scoring three to four require medium-term capability building. Domains scoring five represent strengths that can be leveraged to support capability development in weaker domains. The composite scoring mechanism ensures that remediation resources are directed to the areas of greatest need while recognising that sovereignty improvement is an ecosystem-wide process, not a series of isolated fixes.

The TEE Method provides a structured process for evaluating dependency across all five domains. Each domain assessment produces a score from one to five, generating a composite sovereignty profile that guides remediation prioritisation. Domains scoring below three require immediate intervention. Domains scoring three to four require medium-term capability building. Domains scoring five represent strengths that can be leveraged to support capability development in weaker domains. The composite scoring mechanism ensures that remediation resources are directed to the areas of greatest need while recognising that sovereignty improvement is an ecosystem-wide process, not a series of isolated fixes.

The TEE Method provides a structured process for evaluating dependency across all five domains. Each domain assessment produces a score from one to five, generating a composite sovereignty profile that guides remediation prioritisation. Domains scoring below three require immediate intervention. Domains scoring three to four require medium-term capability building. Domains scoring five represent strengths that can be leveraged to support capability development in weaker domains. The composite scoring mechanism ensures that remediation resources are directed to the areas of greatest need while recognising that sovereignty improvement is an ecosystem-wide process, not a series of isolated fixes.

Keep Reading

Related Articles

LEC Magazine

Join Our Community

Exclusive insights & inspiration

✓

Welcome to LEC!

Account created. Refreshing…

LEC Magazine

Join Our Community

Exclusive insights & inspiration

✓

Welcome to LEC!

Account created. Refreshing…