OAS - AI Governance Framework

Overview

The document, officially titled "Declaration: Towards the Safe, Secure, and Trustworthy Development and Deployment of Artificial Intelligence in the Americas: The Importance of Governance, Regulatory, and Institutional Frameworks," represents a pivotal step by the Organization of American States (OAS) in addressing the burgeoning field of Artificial Intelligence (AI) within the Western Hemisphere. Adopted at the Seventh Meeting of Ministers and High Authorities of Science and Technology (VII REMCYT) on December 12 and 13, 2024, in Washington, D.C., this Declaration, alongside its accompanying Plan of Action, establishes a foundational framework for the responsible evolution and application of AI across the 35 member states of the Americas. Its primary purpose is to guide member states in harnessing AI's transformative potential for integral development, while simultaneously safeguarding human rights, democratic values, and societal well-being. The Declaration underscores the urgent need for a coordinated regional approach to AI governance, recognizing that the opportunities and risks associated with AI transcend national borders and require collective action.

The significance of this instrument lies in its comprehensive scope, aiming to foster an environment where AI innovation thrives responsibly. It acknowledges AI's capacity to drive economic growth, enhance public services, and contribute to the achievement of the 2030 Agenda for Sustainable Development, yet it equally emphasizes the imperative to mitigate risks such as bias, privacy infringement, and misuse. The context of its creation reflects a growing global consensus on the necessity of ethical AI governance, adapted to regional specificities. By providing a common set of principles and a strategic plan for implementation, the OAS seeks to ensure that AI technologies are developed and deployed in a manner that is human-centric, transparent, accountable, and inclusive, thereby building public trust and confidence in these advanced systems across the diverse nations of the Americas.

Definitions

While the Declaration and its accompanying Plan of Action do not present a formal, exhaustive glossary, they implicitly and explicitly define several key concepts through their usage and stated objectives. Central to the document is the notion of "Artificial Intelligence (AI)" itself, understood broadly as a computer science discipline focused on enabling machines to autonomously imitate human intelligence behaviors through models and algorithms. This encompasses the manifestation of intelligence, akin to human beings, within non-human entities. The document's core objective revolves around the "safe, secure, and trustworthy development and deployment of AI," thereby establishing these three adjectives as fundamental definitional pillars for responsible AI.

"Trustworthy AI" is conceptualized as AI systems that are designed, developed, and deployed in a manner that is human-centric, responsible, and respectful of human rights and democratic values, ultimately fostering public trust and confidence. This implies attributes like fairness, explainability, and adherence to ethical guidelines. "Safe AI" refers to AI systems that are developed and deployed with robust mechanisms to identify, understand, and mitigate potential risks and harms, whether physical, psychological, social, or economic, to individuals, society, and the environment. "Secure AI" pertains to the protection of AI systems against malicious actors, cyber threats, and unauthorized access or misuse, ensuring the integrity and resilience of data and algorithms. Furthermore, the document refers to an "Inter-American AI Agenda," which signifies a clear, actionable path for the region to ensure that AI development and governance align with innovation, human rights, and integral development, contributing to an improved quality of life for citizens across the Americas.

Governance and Institutional Framework

The governance and institutional framework for AI within the Organization of American States (OAS) is primarily driven by the mandates stemming from the Meetings of Ministers and High Authorities of Science and Technology (REMCYT) and implemented through the Inter-American Committee on Science and Technology (COMCYT). The VII REMCYT Declaration explicitly urges OAS Member States to engage in a concerted effort to ensure the safe, secure, and trustworthy development and deployment of AI in the Americas. This commitment involves defining and implementing an Inter-American AI agenda, charting a clear path forward with actions that ensure AI development, deployment, and governance are conducted in a manner that fosters innovation, respects, protects, and promotes human rights, and supports integral development for the improvement of citizens' quality of life.

COMCYT, a subsidiary of the Inter-American Council for Integral Development (CIDI), plays a crucial role as the principal forum for hemispheric dialogue and policy formulation in science and technology. It is tasked with taking action on ministerial mandates, undertaking targeted technical cooperation initiatives, and collaborating with strategic partners from the private sector, academia, and non-governmental organizations to promote scientific and technological development in the region. The OAS Executive Secretariat for Integral Development (OAS-SEDI) provides technical secretariat support to COMCYT. The Plan of Action endorsed by the VII REMCYT further outlines the establishment of a Working Group on Governance, Regulatory, and Institutional Models on AI, which is responsible for sharing information, collaborating on model frameworks, and encouraging responsible government design and use of AI. This multi-stakeholder involvement, including civil society, academia, and the private sector, is crucial to ensure that AI policies are comprehensive and reflect diverse perspectives across the Americas.

Key Provisions

The core of the OAS AI Governance Framework for the Americas, as articulated in the VII REMCYT Declaration and its Plan of Action, is built upon a set of guiding values and strategic objectives. Member states reaffirm their commitment to transparency, accountability, inclusivity, security, regional cooperation, and the ethical use of AI. These principles are deemed essential for fostering innovation while simultaneously safeguarding human rights and promoting sustainable development throughout the region. The Declaration specifically urges member states to ensure the safe, secure, and trustworthy development and deployment of AI, committing to define and implement an Inter-American AI agenda that prioritizes human rights, integral development, and an improved quality of life for citizens.

Key provisions include encouraging existing and new OAS Centers of Excellence on Transformative Technologies to conduct research that informs government decisions and policies, provides practical applications, and solves community challenges. Member states are urged to support capacity-building initiatives focused on emerging AI technical and policy developments, including strengthening measurement, intellectual property frameworks, and risk management frameworks. Furthermore, the framework emphasizes enhancing regional AI governance frameworks to foster innovation while respecting human rights and public well-being. It promotes the safe, secure, and trustworthy use of AI through a focus on transparency, accountability, and inclusivity, and aims to strengthen institutional capacities for governance and AI innovation. The Declaration also stresses the importance of aligning AI policies with broader sustainable development goals, encompassing economic growth and social equity across the Americas.

Scope and Application

The OAS AI Governance Framework for the Americas is designed to apply broadly across the 35 member states of the Organization of American States. Its geographic scope is the entire Western Hemisphere, reflecting the OAS's mandate to promote cooperation and integral development among its members. The Declaration and its Plan of Action are intended to serve as guiding instruments for all AI actors within these member states, including governments, private sector entities, academic institutions, civil society organizations, and individual citizens involved in the development, deployment, and use of Artificial Intelligence technologies. The framework's recommendations are non-binding, characteristic of a declaration and plan of action from a ministerial meeting, but they carry significant political and moral weight, encouraging harmonization of approaches and best practices across the region.

The application of these principles is envisioned to be flexible, allowing each member country to develop its specific regulations and policies in alignment with the broader Inter-American framework. The document explicitly calls for supporting inter-sectoral dialogue and multi-stakeholder exchanges to facilitate the pertinence, applicability, and collaborative development of governance, legal, institutional, and regulatory AI frameworks within the Americas. This inclusive approach recognizes the diverse stages of AI development and regulatory maturity among member states, aiming to foster a common understanding and shared commitment to responsible AI while accommodating national contexts and priorities. The ultimate goal is to create a cohesive regional ecosystem where AI can advance equitably and ethically, benefiting all segments of society and addressing challenges that transcend national borders.

Implementation Framework

The implementation framework for the OAS AI Governance Principles is primarily outlined in the Plan of Action endorsed by the VII REMCYT, which serves to guide the execution of priorities established in the Declaration. A central component of this framework is the establishment of a Working Group 1 on Governance, Regulatory, and Institutional Models on AI. This working group is tasked with critical activities such as sharing information for the development of model AI policies and frameworks, including good practices, lessons learned, case studies, and comparative models. The objective is to build a common baseline of understanding on AI for the Americas, facilitating a harmonized regional approach. The working group will also collaborate in designing model governance, regulatory, and institutional frameworks for the safe, secure, and trustworthy development and deployment of AI.

Beyond the working group, the implementation framework emphasizes several key strategies. These include supporting regional collaboration on AI capacity-building initiatives with COMCYT partners in governments, academia, the private sector, and civil society. Coordination with other regional organizations, development banks, and partners guiding AI initiatives with OAS member states is crucial to identify collaboration opportunities, maximize resources, and avoid duplication of efforts. Member states are encouraged to promote responsible government design, development, and use of AI that respects the protection and promotion of human rights and fundamental freedoms. Furthermore, the framework calls for supporting inter-sectoral dialogue and multi-stakeholder exchanges to facilitate the pertinence, applicability, and collaborative development of governance, legal, institutional, and regulatory AI frameworks in the Americas, ensuring a comprehensive and inclusive approach to AI integration.

Monitoring and Evaluation

The Declaration and its Plan of Action implicitly establish mechanisms for monitoring and evaluation through their emphasis on ongoing collaboration, information sharing, and the continuous development of model frameworks. While a specific, standalone monitoring and evaluation section is not explicitly detailed, the nature of the Inter-American AI agenda and the functions assigned to the Inter-American Committee on Science and Technology (COMCYT) and its working groups suggest an iterative process of review and adaptation. COMCYT, as the primary forum for hemispheric dialogue on science and technology, is inherently positioned to oversee the progress of ministerial mandates, including those related to AI governance.

The Plan of Action encourages member states to share information on the status quo of AI policy and the regulatory environment throughout the region. This continuous exchange of data and experiences is vital for assessing the effectiveness of implemented policies and identifying areas requiring further attention or adjustment. The call for coordination with other regional organizations and partners also implies a benchmarking process, where the OAS can learn from and contribute to broader international efforts in AI governance. By fostering an environment of transparency and open dialogue, the framework aims to enable member states to collectively evaluate the impact of AI initiatives, ensuring that they remain aligned with the guiding values of human rights, sustainable development, and innovation, and are responsive to the evolving challenges and opportunities presented by AI technologies.

Relationship to Other Instruments

The OAS AI Governance Framework for the Americas is explicitly situated within a broader landscape of existing international and regional instruments, building upon previous commitments and aligning with global best practices. The Declaration recalls the Declaration of Jamaica, titled “Harnessing the Power of Science and Transformative Technologies to Drive our Communities Forward,” which was adopted at the Sixth Meeting of Ministers and High Authorities of Science and Technology (VI REMCYT) in 2021. This earlier declaration encouraged member states to promote the use of science, technology, and innovation for integral development and to deepen regional collaboration to close technological and social gaps, particularly for vulnerable populations. The current AI framework thus represents a direct evolution and specific application of these prior commitments to the domain of Artificial Intelligence.

Furthermore, the Plan of Action explicitly states its intention to guide the implementation of priorities established in the VII REMCYT Declaration and to contribute to the achievement of the 2030 Agenda for Sustainable Development. This demonstrates a clear link to the United Nations' global sustainable development goals, positioning responsible AI development as a tool for achieving broader societal objectives. The framework also implicitly acknowledges and seeks alignment with other international efforts, such as the OECD AI Principles, by promoting similar values like transparency, accountability, and human-centric design. By referencing and integrating these various instruments, the OAS AI Governance Framework aims to ensure coherence, avoid duplication, and leverage existing global and regional expertise in its approach to AI governance, fostering a harmonized and effective strategy across the Americas.

International Alignment

The OAS AI Governance Framework for the Americas demonstrates a strong commitment to international alignment, recognizing that Artificial Intelligence is a global phenomenon requiring coordinated responses. The Declaration and its Plan of Action are grounded in shared democratic values and emphasize regional cooperation, which inherently positions the framework within the broader international discourse on AI governance. The document explicitly states that it seeks to unify member states under shared values like transparency, accountability, inclusivity, security, cooperation, and ethical use of AI to ensure coherence and effectiveness in AI governance, thereby contributing to global efforts for responsible AI.

While not explicitly naming every international instrument, the principles articulated in the OAS framework resonate strongly with those found in other prominent global AI governance initiatives, such as the OECD AI Principles and the UNESCO Recommendation on the Ethics of Artificial Intelligence. The emphasis on human rights, non-discrimination, privacy, safety, and accountability reflects a common understanding of ethical AI development that is gaining traction worldwide. The framework's call for coordination with other regional organizations and multilateral institutions further underscores its intent to contribute to and benefit from a global network of AI governance efforts. By fostering an environment of information sharing and collaborative development of model policies, the OAS aims to ensure that the Americas' approach to AI is not only regionally relevant but also internationally compatible and supportive of a safe, secure, and trustworthy global AI ecosystem.

Implementation Timeline

Adoption and Endorsement

Sources and References