NetSwitch Limited

1. Clear rules, not theories. A few hard, usable standards on safety, transparency, and accountability something countries can adopt immediately. 2. Real commitments. Not "we agree in principle." Actual timelines, signed participation, and consequences for non-compliance. 3. Inclusion with power. Emerging economies (like Nigeria) must leave with funding, access to infrastructure, and talent support not just a seat at the table. 4. Innovation still alive. Regulation that protects people but doesn't choke startups think sandboxes and phased rules. 5. Crisis readiness. A global system to report and respond to AI failures fast. If something breaks, the world shouldn't scramble.

AI capacity-building;Open-source software, open data and open AI models;

Open-source software, open data, and open AI models lower the barriers to entry, allowing countries like Nigeria to build, adapt, and deploy AI solutions without relying entirely on expensive foreign technologies. They also enable local talent to learn, innovate, and create context-specific solutions, ensuring participation in the global AI economy rather than remaining just consumers.

One major gap is the growing concentration of AI power in a few companies controlling compute, data, and models, which risks locking out emerging economies from meaningful participation. Another is the lack of focus on local context without culturally relevant data and governance input, AI systems will reinforce bias and fail in markets like Nigeria.

Nigeria is mainly affected by weak AI governance through dependence on foreign AI systems and limited control over how data is used, which creates risks of bias, misuse, and poor local relevance. At the same time, gaps in infrastructure and regulation slow down local innovation and keep most value creation outside the country. However, there is a strong opportunity to leapfrog by using open-source AI and building local datasets, allowing Nigeria to develop more relevant solutions in sectors like finance, agriculture, and public services.

The AI Dialogue can act as a bridge for aligning countries on shared minimum standards for AI safety, transparency, and accountability, reducing fragmentation in global governance. It can also create trust and coordination mechanisms that help countries share knowledge, resources, and best practices, especially supporting emerging economies to participate meaningfully in AI development.

Existing efforts like the UNESCO's AI Ethics framework, and the Global Partnership on AI already provide useful foundations for shared norms and ethical guidance. The AI Dialogue can add value by turning these principles into practical, coordinated actions especially by linking them to implementation, funding, and inclusion for countries that are currently underrepresented in AI development.

Governments can set baseline rules and commit to implementation, while industry provides technical expertise, infrastructure, and responsible deployment practices. Academia and civil society should stress-test policies, flag risks, and ensure public interest is not sidelined. For structure, the Dialogue should be outcome driven, not speech-driven small working groups focused on specific deliverables, clear timelines, and published commitments rather than open-ended discussions.

Communities in the Global South, including African countries like Nigeria, are underrepresented, along with local technologists, informal sector workers, and communities most affected by AI deployment rather than its design. They can be included through structured participation in decision making forums, funded representation, and open channels for local case studies and lived experiences to directly shape policy, not just comment on it.

Small, focused working groups that produce specific outputs (like draft standards or pilots) would drive more value than large plenary sessions. Alongside that, live policy sandboxes and real world case demonstrations would make discussions practical, letting stakeholders test governance ideas in action rather than just debating them.

The EU AI Act is a strong example of a risk based regulatory approach that links rules to the level of potential harm, not just broad principles. On the practical side, AI "sandboxes" used in places like the UK and Singapore allow companies to test systems under regulator supervision, balancing innovation with safety.