NIGERIA FORUM FOR AI GOVERNANCE

A successful inaugural Dialogue must achieve four critical outcomes: Equitable Representation: Transitioning from consumer to co-driver status for Global South nations, ensuring international standards are not solely dictated by major tech powers. Resource Interoperability: Establishing concrete mechanisms for equitable access to high-performance computing (HPC) and inclusive datasets to bridge the "AI divide". Interoperable Standards: Facilitating a baseline for "Secure-by-Design" AI that respects diverse cultural and linguistic contexts while preventing regulatory arbitrage. Tangible Commitments: Moving beyond aspirational principles to actionable, binding implementation plans for the Global Digital Compact, including clear pathways for national domestication.

• AI capacity-building
• Safe, secure and trustworthy AI
• Social, economic, ethical, cultural, linguistic and technical implications of AI
• Protection and promotion of human rights

These selections align with Nigeria's National Artificial Intelligence Strategy (NAIS). Capacity-building is critical for our youth-led economy, exemplified by the 3 Million Technical Talent (3MTT) program by our national government. Safe/Trustworthy AI is essential to mitigate domestic risks, such as data poisoning and fraud, in our rapidly digitizing banking and healthcare sectors. Socio-cultural/Linguistic implications are prioritized to foster localized innovation, such as government-backed Large Language Models (LLMs) for Hausa, Igbo, and Yoruba. Human Rights ensures that AI adoption does not exacerbate existing inequalities or facilitate unethical data extraction".

The NFAIG identifies Digital Sovereignty and Fair Data Compensation as a vital emerging issue. This includes the right of nations to require that data generated by their citizens contribute directly to the local economy. For instance, Nigeria's proposed Digital Sovereignty and Fair Data Compensation Bill introduces a 2% National AI Development Fund contribution from foreign firms using local data, addressing the unregulated extraction of value from the Global South.

Challenges: Africa faces regulatory fragmentation, with inconsistent data protection regulations and emerging AI policies, leading to uncertainty for innovators. There is also a severe infrastructure gap regarding GPU/TPU availability and domestic cloud hosting. Opportunities: The Nigeria National AI Strategy 2025 and the upcoming National Digital Economy and E-Governance Bill provide a window to institutionalize a risk-based, multistakeholder approach. The NFAIG serves as the bridge to harmonize these efforts into a unified national posture.

The Dialogue should serve as a "Dialogue of Dialogues," connecting disparate regional initiatives (like the AU Continental AI Strategy) to promote mutual recognition of standards. It can facilitate a Global Frontier AI Governance Evaluation Framework, as a common assessment environment that allows developing nations to consult global stakeholders, engage peers, and validate the safety of foreign models before large-scale domestic deployment.

The Dialogue should build upon the UNESCO Recommendation on the Ethics of AI, the OECD AI Principles, and the Bletchley Declaration, and Internet Governance outcomes on AI ethics. Added value lies in its universal home at the UN, which grants it the legitimacy to host an Independent International Scientific Panel on AI that separates scientific evidence from short-term market pressures.

NFAIG recommends a Multistakeholder Advisory Group (MAG) modeled on the Internet Governance model, with rotating leadership among government, the private sector, and civil society. This ensures that "seat at the table" translates into agenda-setting power. We suggest structured thematic working groups that meet intersessionally to produce evidence-based assessments for the annual plenaries.

Global South innovators, local language researchers, and civil society organizations (CSOs) in developing markets are underrepresented. They can be included through Participation Support funds, regional "hubs" such as the NFIAG recently established by Global Network for Cybersolution that feed into the Global Dialogue, and by mandating that a percentage of the Dialogue's leadership roles be held by representatives from emerging AI economies

While panel and other contemporary formats are good, the UN should avoid the usual talk-show format. The engagement should be impactful, with measurable, evidence-based outcomes. For example, hybrid digital sandboxes allow cross-border assessment and formulation of global policies and products under joint multistakeholder oversight. AI for Good Hackathons that co-locate rapid problem-solving prototyping events with dialogue to showcase localized SDG solutions should be considered.

A rotating secretariat model that prevents institutional capture and ensures balanced ecosystem growth is paramount. Another is the Nigeria Data Protection Commission (NDPC) Annual Compliance Audits, which use the NDPA 2023 as a benchmark for responsible data usage in AI workflows and mandate that organizations in Nigeria implement this practice. This practice is working and practically effective. A similar practice model can be adopted. To promote effective AI governance, particularly within the framework of a dynamic Governance Model, the focus should shift from static regulation to dynamic, adaptive policies. One of our members developed the "REST AI Governance Framework" and its subsequent expansion into a white paper, made possible by the invaluable support, insights, and early reviews of distinguished experts across the fields of artificial intelligence, cybersecurity, governance, and policy. Perhaps a copy can be downloaded to explore its consideration for an Adaptive and Trusted AI Governance Framework for the Global South.