NEO CONSULTING S.A.C.

From my position as a Senior Data Scientist and as a participant in various conferences and discussion forums on Internet governance and AI governance, I believe that the main indicator of success for the first Global Dialogue on AI Governance would be the ability to move from conceptual discussions toward concrete implementation mechanisms. Too often, these spaces focus on defining what AI governance is, what ethical principles should guide it, and what the associated risks are. However, there is still very limited discussion on how countries, organizations, and technical teams can realistically implement these principles within their processes, technological architectures, and operating models. It would be valuable for this dialogue to generate actionable guidelines, methodologies, standards, and adoption frameworks capable of translating AI governance into real-world environments with different levels of technological maturity. I also believe it is important to broaden the conversation beyond Generative AI. Currently, much of the AI governance debate is dominated by the rise and media attention surrounding GenAI, while other branches of artificial intelligence require equal or even greater attention. Machine Learning, for example, requires both Machine Learning Operations (MLOps) and effective governance, since these systems often work directly with private or sensitive personal information and participate in automated decision making processes. This creates challenges related to privacy, traceability, explainability, algorithmic bias, and model lifecycle control. Another important aspect is inclusion and education in AI. We are living in a world where children and younger generations are becoming increasingly technological, while many school curricula continue to focus primarily on traditional office software skills. In this context, it is essential to promote AI literacy and inclusion in order to ensure a responsible, critical, and informed use of these technologies from an early age. Therefore, a true success of this dialogue would be to build a more comprehensive, technical, and applicable vision of AI governance, one that considers the entire artificial intelligence ecosystem and not only the technologies currently receiving the highest level of media attention.

• Transparency, accountability, and human oversight
• AI capacity-building
• Social, economic, ethical, cultural, linguistic and technical implications of AI
• Interoperability of governance approaches

I selected these priorities because, from my experience, one of the main gaps in current conversations is the lack of focus on practical implementation. Many governance discussions remain at the conceptual or ethical level, while there is still limited attention on how organizations, governments, and technical teams can operationalize governance principles through standards, processes, and technical frameworks that are adaptable across different contexts and levels of technological maturity. In this regard, interoperability of governance approaches is essential to help organizations adopt governance mechanisms that are practical, scalable, and aligned across different technological and regulatory environments. At the same time, transparency, accountability, and human oversight are critical to ensuring that AI systems remain understandable, traceable, and properly supervised, especially in Machine Learning environments where automated decision making and model lifecycle management can directly impact people and organizations. I also consider AI capacity building to be fundamental, particularly in a context where younger generations are increasingly exposed to AI technologies while many educational systems still focus mainly on traditional digital and office software skills. Promoting AI literacy and inclusion is necessary to ensure responsible, informed, and critical use of these technologies from an early age. In the same way, it is also necessary to include older generations in this transition to avoid widening existing digital gaps. For this reason, it is essential to consider the social, economic, ethical, cultural, linguistic, and technical implications of AI, ensuring that governance frameworks are inclusive and capable of addressing the diverse realities and needs of different populations.

One important emerging topic is the operationalization of AI governance. Many discussions focus on principles, ethical guidelines, and high level governance frameworks, but there is still limited attention on how organizations and governments can implement these principles in practice through technical standards, monitoring frameworks, MLOps processes, model lifecycle management, and measurable governance mechanisms adaptable to different levels of technological maturity.

From my experience as a Senior Consultant developing AI services for many of the leading companies in Peru across different sectors, one of the main challenges is that most organizations still lack mature data governance frameworks and, in many cases, AI governance structures are almost non existent. This has created significant concerns regarding information leakage, improper use of AI tools, accidental data deletion, unreliable outputs, and unexpected operational or infrastructure costs. As a result, many organizations remain cautious even when adopting relatively basic AI solutions. At the same time, Peru has made important progress through Law No. 31814 and its regulatory framework, which establish principles for ethical, transparent, and responsible AI use. The regulation represents an important opportunity to promote accountability, risk management, transparency, and protection of fundamental rights in the use of AI systems. However, one of the main challenges is that the regulatory approach may advance faster than the country's actual technological maturity and institutional readiness. Some organizations, startups, universities, and small businesses may face difficulties complying with requirements related to impact assessments, technical documentation, traceability, audits, and specialized AI governance roles, particularly in environments where AI adoption is still emerging. For this reason, one of the greatest opportunities for Peru and the region is to achieve a better balance between regulation and innovation by strengthening digital infrastructure, AI literacy, technical talent development, and practical governance implementation capabilities. Additionally, from my experience as an AI educator, I have observed that both children and older adults are among the groups most disconnected from these discussions, highlighting the importance of inclusion and accessible AI education as part of long term governance strategies.

The AI Dialogue can play a key role in fostering international cooperation by helping countries move beyond isolated regulatory approaches and toward more interoperable, practical, and inclusive governance frameworks. One of the main current challenges is that many countries, particularly in developing regions, are attempting to adopt governance models inspired by more technologically mature economies without having the same infrastructure, institutional capacity, or level of AI adoption. In this context, the Dialogue could help promote the exchange of practical experiences, technical standards, governance methodologies, and implementation strategies that are adaptable to different realities and levels of technological maturity. This is especially important for regions such as Latin America, where many organizations are still in early stages of AI adoption and governance development. The Dialogue can also contribute to reducing global AI inequalities by encouraging international collaboration in areas such as AI literacy, talent development, infrastructure access, research cooperation, and knowledge sharing. Governance discussions should not focus only on regulation, but also on enabling countries and organizations to build the technical and institutional capabilities necessary to implement governance effectively. Therefore, the Dialogue provides an important opportunity to ensure that AI governance discussions include perspectives from developing countries, educators, technical communities, startups, and civil society, allowing governance frameworks to become more globally representative, inclusive, and applicable across diverse contexts.

The AI Dialogue should build upon existing initiatives and governance efforts such as the OECD AI Principles, UNESCO's Recommendation on the Ethics of Artificial Intelligence, the Global Partnership on AI (GPAI), Internet Governance Forum (IGF) discussions, and emerging regulatory frameworks such as the EU AI Act. These initiatives have already contributed valuable principles, ethical guidelines, and policy discussions that can serve as an important foundation for international cooperation. At the same time, the Dialogue could add significant value by helping connect these high level frameworks with practical implementation needs, particularly for developing countries and organizations with lower levels of technological maturity. In many cases, existing initiatives provide broad governance principles but less guidance on operationalization, technical implementation, organizational adoption, and measurable governance mechanisms. The Dialogue could also serve as a bridge between policymakers, technical communities, academia, educators, startups, and the private sector to ensure that governance discussions are not limited to regulatory perspectives alone. This is particularly important in regions such as Latin America, where AI adoption is growing rapidly but governance capabilities, infrastructure, and specialized talent are still developing. Additionally, the Dialogue can help promote more globally inclusive governance discussions by incorporating perspectives from countries and sectors that are often underrepresented in international AI policy conversations. This would contribute to governance frameworks that are more adaptable, realistic, and applicable across diverse economic, social, and technological contexts.

Different stakeholders can contribute to the AI Dialogue by bringing complementary perspectives and practical experiences. Governments can contribute regulatory and public policy perspectives, while the private sector and technical communities can provide insights into implementation challenges, operational governance, and emerging technological risks. Academia and educators can support research, AI literacy, and long term capacity building, while civil society organizations can help represent the social impact of AI on different populations. To make participation more effective, the Dialogue should combine high level policy discussions with technical and implementation focused sessions. Including workshops, case studies, regional working groups, and multi stakeholder roundtables would help create more practical and actionable outcomes instead of remaining only at the conceptual level. It would also be valuable to maintain hybrid and multilingual participation formats in order to include voices from developing countries and communities with more limited access to international governance spaces.

One of the most underrepresented perspectives in global AI governance discussions is that of developing countries and regions with lower levels of technological maturity, particularly in Latin America and other emerging economies. In many cases, global governance discussions are dominated by countries and organizations with greater technological infrastructure, funding, and regulatory capacity. Technical practitioners working directly on AI implementation, educators, startups, small organizations, and local innovation ecosystems are also often underrepresented, despite facing many of the real operational challenges associated with AI adoption and governance. Additionally, children, older adults, and populations with limited digital literacy are rarely included in discussions, even though they are among the groups most impacted by the digital divide and the rapid expansion of AI technologies. These perspectives could be better included through multilingual participation, regional consultation mechanisms, scholarships or funding support for underrepresented participants, hybrid participation formats, and more direct collaboration with universities, technical communities, educators, and civil society organizations.

More innovative participation formats could help make the AI Dialogue more dynamic, practical, and inclusive. In addition to traditional panel discussions, the Dialogue could include technical workshops, collaborative problem solving sessions, governance simulation exercises, regional innovation labs, and real world case study discussions focused on practical implementation challenges. Interactive multi stakeholder roundtables could also help create more balanced discussions between policymakers, technical experts, educators, civil society, startups, and private sector representatives. This would allow participants to move beyond theoretical debates and work collaboratively on governance approaches adapted to different realities and levels of technological maturity. It would also be valuable to incorporate hybrid and asynchronous participation mechanisms, such as digital consultation platforms, open collaborative documents, and regional online sessions before the main Dialogue. This would help include participants who may not have the resources or geographic access to attend international events in person.

Some practical approaches that can promote effective AI governance include establishing clear role and permission structures for cloud platforms and AI systems, ensuring that only authorized personnel can access sensitive information or modify critical infrastructure. It is also important to designate project owners or governance leaders with a holistic understanding of technical, commercial, social, and ethical implications in order to balance innovation with responsible use. From an operational perspective, organizations should implement budget alerts, monitoring systems, secure API key management practices, audit logs, and governance mechanisms for datasets and model lifecycle management. In Machine Learning environments, maintaining representative and balanced datasets is essential to reduce bias and discrimination risks in automated decision making systems. Data governance practices are also critical to ensure data quality, integrity, traceability, and proper usage during model training and deployment. Another important practice is the incorporation of human oversight mechanisms for high impact AI systems, particularly in sectors such as healthcare, finance, education, and public services. Regular risk assessments, explainability evaluations, and monitoring of model performance over time can also help improve accountability and trust. From an educational and social perspective, AI literacy should be incorporated into school curricula as a mandatory component of modern digital education. In parallel, governments, universities, and private organizations should promote accessible AI training programs for vulnerable communities, children, older adults, and people without technical or university education in order to reduce digital inequalities and encourage more inclusive participation in the AI ecosystem.