ALGERIE TELECOM

The success of the first Global Dialogue on Artificial Intelligence Governance will depend on its ability to produce concrete, inclusive and sustainable results. First of all, it would be essential to achieve a clear international governance framework, promoting an ethical, responsible, and secure use of AI. This framework should be flexible enough to adapt to rapid technological developments, while ensuring the protection of fundamental rights. Secondly, the establishment of a collaborative global platform, under the supervision of bodies such as the International Telecommunication Union (ITU), would represent a major step forward. Such a platform, inspired by existing models like Hugging Face, would allow for the centralization and sharing of AI models, datasets and best practices, while ensuring transparency, traceability and equitable access, particularly for developing countries. Furthermore, the Dialogue should promote skills transfer and capacity building in order to reduce the digital divide and allow all states to benefit from the opportunities offered by AI. Finally, success will be measured by the creation of effective international cooperation mechanisms, including concrete commitments, monitoring tools and impact indicators. The active involvement of governments, the private sector, academia and civil society will be crucial to ensure inclusive governance. Thus, a successful dialogue will be one that transforms discussions into tangible actions, in the service of sustainable, equitable and human-centered development.

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

The two priorities selected - AI capacity building and - open source software, open data and open AI models are closely linked and constitute essential levers for an inclusive and sustainable development of artificial intelligence. AI capacity building is fundamental to reduce digital inequalities between countries and enable equitable participation in the AI economy. This involves the development of technical skills, training of local talent, and access to necessary infrastructure and resources. Without this investment, many countries risk remaining mere consumers of technologies rather than innovative players. At the same time, the promotion of open source software, open data and open AI models plays a decisive role in the democratization of AI. These approaches promote transparency, collaborative innovation and equitable access to technologies. They also help to strengthen the digital sovereignty of States and encourage the emergence of solutions adapted to local contexts. In this perspective, the creation of an international collaborative platform, inspired by Hugging Face and placed under the aegis of the International Telecommunication Union, could constitute a structuring initiative. Such a platform would facilitate the secure sharing of models, datasets and best practices, while supporting skills development on a global scale. Thus, the articulation between human capacities and technological openness represents a key condition for more equitable, inclusive, and sustainable AI governance.

Yes, several cross-cutting and emerging issues deserve particular attention, as they complement the existing themes and condition effective and sustainable governance of AI. First of all, digital sovereignty and infrastructure control are a major strategic issue. Beyond models and data, dependence on infrastructure (cloud, compute-intensive, GPU) can accentuate global imbalances. AI governance must therefore integrate equitable access to computational resources and encourage regional infrastructures. Then, the issue of AI's environmental sustainability becomes critical. The training and deployment of models consume significant energy resources. It is necessary to promote 'green' AI standards, integrating carbon footprint indicators and energy optimization practices. Furthermore, the governance of sensitive and strategic data (notably public, linguistic or cultural) represents a cross-cutting issue. It is important to ensure their protection while allowing their responsible valorization, in order to avoid an unbalanced exploitation or a loss of cultural diversity. Another emerging dimension concerns the evaluation and certification of AI models. The lack of harmonized mechanisms to test, audit and certify AI systems limits trust and comparability between solutions. The establishment of international benchmarks would be a structuring lever. Finally, linguistic inclusion, particularly for underrepresented languages such as Arabic or African languages, remains insufficiently addressed. Ensuring their presence in AI models is essential for fair adoption. Thus, integrating these transversal dimensions would significantly strengthen the relevance and impact of the Global Dialogue on AI Governance.

In our current context, governance gaps in AI capacity building and open models promotion have a direct impact on our ability to fully leverage the opportunities offered by artificial intelligence. Among the main challenges, there is first a lack of specialized skills and limited access to advanced computing infrastructures. This hinders the local development of innovative solutions and reinforces dependence on imported technologies. Moreover, the lack of clear frameworks for sharing and enhancing data limits the exploitation of strategic resources, particularly in the public and economic sectors. Another challenge lies in the weak structuring of open source and data ecosystems. Although initiatives exist, they remain fragmented and lack coordination at national and regional levels. This reduces opportunities for collaboration, collective innovation and resource sharing. However, these gaps also represent important opportunities. The development of public policies oriented towards open data and open models can accelerate local innovation, strengthen digital sovereignty, and promote the emergence of startups and solutions tailored to the specific needs of the territory. In this regard, drawing inspiration from platforms like Hugging Face and promoting initiatives under the auspices of international organizations such as the International Telecommunication Union would allow for the structuring of a collaborative, inclusive, and sustainable ecosystem. In sum, filling these governance gaps is an essential condition to transform the current constraints into levers of economic, technological and social development.

The Global Dialogue on AI Governance can play a structuring role as a platform for international convergence, facilitating coordination between states, international organizations, the private sector and academia. First of all, it can contribute to harmonizing governance approaches by promoting the development of common principles and international standards. This convergence is essential to avoid regulatory fragmentation and ensure consistent, ethical and secure use of AI globally. Secondly, the Dialogue can act as a catalyst for technical cooperation and knowledge sharing. By encouraging the exchange of good practices, feedback and expertise, it would strengthen the capacities of countries, particularly those in developing countries, and reduce technological gaps. Furthermore, it can support the implementation of concrete collaborative initiatives, such as international platforms for sharing models, data, and AI tools. In this regard, drawing inspiration from initiatives such as Hugging Face, under the coordination of bodies like the International Telecommunication Union, would make it possible to structure an ecosystem that is open, transparent and accessible to all. The Dialogue can also play a key role in mobilising resources and partnerships by facilitating funding for joint projects, particularly in the fields of training, research and innovation. Finally, its success will be based on its ability to transform exchanges into operational commitments, accompanied by monitoring and evaluation mechanisms. Thus, the Dialogue on AI can become a central lever for building an inclusive, balanced and sustainable global governance.

The Global Dialogue on AI Governance should build on several existing initiatives and frameworks to avoid duplication of efforts and strengthen international coherence. Among the key references are the work of the Organisation for Economic Co-operation and Development on AI principles, those of UNESCO related to the ethics of artificial intelligence, as well as the technical initiatives led by the International Telecommunication Union in terms of standardization and connectivity. Multi-stakeholder partnerships such as the Partnership on AI also play an important role in promoting good practices. In parallel, collaborative platforms such as Hugging Face illustrate the potential of open ecosystems for sharing models and data, while initiatives like AI for Good demonstrate the impact of AI in sustainable development. The added value of the AI Dialogue lies in its ability to bring together these initiatives within a coherent global framework, ensuring better coordination between political, technical and operational dimensions. Unlike existing initiatives, often sectoral or regional, the Dialogue can offer an inclusive space at the global level, integrating the priorities of developing countries. It could also act as a bridge between principles and implementation, translating existing frameworks into concrete actions, accompanied by mechanisms for monitoring, evaluation and impact sharing. Finally, the Dialogue can catalyze the creation of new structuring initiatives, including open international platforms for AI models and data, thus strengthening cooperation, transparency, and innovation on a global scale.

The success of the AI Dialogue is based on an active and structured participation of all stakeholders, each bringing complementary value. Governments can contribute by defining regulatory frameworks, sharing national experiences and supporting public policies that promote responsible innovation. The private sector plays a key role in technological development, providing technical expertise and sharing operational best practices. The academic world and research centers provide essential scientific expertise, particularly in terms of evaluation, ethics, and impact. Finally, civil society guarantees an inclusive approach, ensuring that human rights and social and cultural issues are taken into account. To maximize its impact, the Dialogue should adopt a hybrid format, combining strategic plenary sessions and thematic working groups oriented towards concrete results. These groups could be organized around key priorities such as capacity building, open models or data governance. It is also recommended to integrate digital participatory mechanisms, allowing for continuous contribution (collaborative platforms, open consultations), in order to broaden participation beyond physical events. Structurally, the Dialogue would benefit from including: - clear and measurable objectives per cycle, - concrete deliverables (recommendations, standards, roadmaps), - a monitoring and evaluation mechanism, - and a permanent secretariat to ensure continuity. Finally, the integration of collaborative tools inspired by platforms like Hugging Face, under the coordination of bodies such as the International Telecommunication Union, would strengthen the operational and inclusive dimension of the Dialogue. Thus structured, the AI Dialogue could become a real driver of international cooperation and concrete action.

In the first place, SMEs, local startups and independent innovators from developing countries, particularly in Africa, are still insufficiently represented and have few opportunities to express themselves in international forums, despite their key role in innovation on the ground. Young emerging talents, students and researchers are also insufficiently integrated into decision-making processes. While they face specific challenges related to infrastructure, skills and data access. Similarly, local communities and end users are rarely involved, although they are directly impacted by AI systems. Underrepresented languages and cultures, particularly African and Arabic languages, also constitute a significant blind spot. This situation influences the quality and fairness of AI models, often biased towards dominant linguistic contexts. To remedy these imbalances, several actions are recommended. It is essential to implement inclusive participation mechanisms, including open consultations, accessible digital platforms and participation support programs (funding, translation, remote access). The Dialogue could also promote dedicated regional spaces, allowing for the expression of local priorities, as well as targeted capacity building initiatives. Truly effective AI governance must imperatively integrate this diversity of voices to reflect global realities.

To foster meaningful and dynamic engagement, the AI Dialogue should adopt innovative, interactive and action-oriented formats. First of all, the organization of collaborative laboratories (AI Policy Labs) would make it possible to bring together decision-makers, technical experts, and field players to co-construct concrete solutions based on real cases. These practical sessions would promote active participation rather than purely theoretical exchanges. Then, open and continuous participatory digital platforms, inspired by Hugging Face, could allow stakeholders to contribute before, during, and after the sessions. These spaces would facilitate the sharing of models, ideas, documents and best practices, while ensuring transparency and traceability. Digital citizen consultations (surveys, open forums, public contributions) also represent an important lever to integrate the perspectives of end users and civil society at large scale. Furthermore, formats such as 'policy hackathons' or innovation challenges could mobilize young talents, startups and البحثين around specific issues of AI governance, with concrete results at stake (prototypes, recommendations, tools). The Dialogue could also integrate immersive and demonstrative sessions, allowing to present real use cases and their impacts (positive and negative), in order to facilitate the understanding of the issues by all participants. Finally, the establishment of hybrid and continuous working groups, supported by organizations such as the International Telecommunication Union, would ensure the continuity of exchanges and follow-up on recommendations. These formats would contribute to transforming the Dialogue into a living, inclusive and results-oriented space.

Several existing policies, initiatives and platforms offer concrete examples of effective AI governance and can inspire the Global Dialogue. At the normative level, the AI Principles of the Organisation for Economic Co-operation and Development constitute an international reference for a responsible, human-centered AI. Similarly, the UNESCO Recommendation on the Ethics of AI proposes a global framework integrating human rights, cultural diversity and transparency. At the regulatory level, the European Union's initiative with the AI Act illustrates a structured approach based on the level of risk, offering a concrete model to frame the uses of AI while supporting innovation. In terms of platforms, Hugging Face represents an emblematic example of an open ecosystem that promotes the sharing of models, data and tools, while encouraging transparency and global collaboration. This approach contributes to democratizing access to AI and accelerating innovation. Furthermore, initiatives such as AI for Good, led by the International Telecommunication Union, demonstrate how AI can be mobilized to respond to concrete challenges related to sustainable development. Finally, some emerging practices, such as algorithmic audits, regulatory sandboxes and impact assessment frameworks, offer operational tools to test, frame and improve AI systems in real-world conditions. These examples show that effective AI governance relies on a balanced combination of principles, regulation, open innovation and international cooperation.