91�鶹������Ʒ����

Fast-paced and ground-breaking, the progress of Artificial Intelligence (AI) permeates many aspects of our lives. AI has enormous potential for generating social welfare and for accelerating human progress and sustainable development. At the same time, the rapid development of AI raises ethical concerns including debates on how to deal with the perpetuation of bias and inequalities, access, capacity development and education on AI, freedom of expression, privacy, transparency, accountability, data disclosure, ownership and access to data, amongst others.

The rapid introduction of AI in the Caribbean sub-region is of no exception. Its uptake gives it a great potential to address some of the Caribbean’s most pressing development challenges, and supporting Small Islands Developing States in the implementation of the SAMOA Pathway. Nevertheless, the increased use of AI-enabled systems and technology may cause disruptive effects on the Caribbean labour market and imperil social cohesion. For this reason, developing a joint normative instrument that outlines the ethics of AI is of importance so that Caribbean SIDS can positively benefit from AI.

In response to these challenges, 91�鶹������Ʒ���� embarked on a two-year process to elaborate the first global standard-setting instrument on the Ethics of AI in the form of a Recommendation, following the decision of UNESCO’s General Conference in November 2019. An Ad Hoc Expert Group (AHEG) composed of 24 international experts appointed by the Director-General of UNESCO, prepared the first in May 2020. In order to craft an inclusive instrument that is anchored in the needs and potentials of different stakeholder groups from across the world, UNESCO in close collaboration with its Member States and further partner organizations, conducted a global consultation process on the draft Recommendation.

The Caribbean Sub-regional Consultation on the draft Recommendation on the Ethics of AI, which took place on 5 August 2020, was hosted by the Government of Barbados with the special participation of the Hon. Mia Mottley, Prime Minister of Barbados.

Hon. Mia Mottley strongly expressed Barbados’ support in defining ethical principles of AI and calling for collaborative efforts towards making use of AI as a catalyst for Agenda 2030 in Caribbean SIDS. She emphasized the importance of embedding the discussion of ethics of Artificial Intelligence in a set of common values and principles, which serve for the benefit of humanity.

Alissandra Cummins, Chairperson of the Barbados National Commission for UNESCO and former Chairperson of the UNESCO Executive Board, facilitated the Caribbean Consultation, with the support of the UNESCO Cluster Office for the Caribbean. The consultation brought together over 20 experts and stakeholders from the English and Dutch-speaking Caribbean Small Island Developing States (SIDS) pertaining to civil society, the public and private sector, as well as ethicists, media, academia, and youth representatives.

In her introductory remarks, Gabriela Ramos, UNESCO Assistant Director-General, Social and Human Sciences Sector advocated for a value-based and humanistic approach to AI as a global compass on steering Artificial Intelligence within the framework of Human Rights.

Guy Berger, Director for Freedom of Expression and Media Development, presented UNESCO’s role in Artificial Intelligence for sustainable development and in creating open knowledge societies. UNESCO as the UN lead agency for Access to information and  knowledge and Ethical dimensions of the Information Society published the report “” which applies UNESCO’s Internet Universality ROAM framework to assess how AI and advanced ICTs impact Human Rights, Openness and Access, and how a Multi-stakeholder approach is necessary to address both the challenges and opportunities.

The draft Recommendation on the Ethics of Artificial Intelligence was presented by Dr. Edson Prestes, Professor at the Federal University of Rio Grande do Sul, Brazil, in his capacity as member of UNESCO’s AHEG. While there exist multiple definitions of AI, he noted that no definition has been able to encompass the richness and complexity of the technology and its ecosystem. The responsibility and accountability for errors and biases in AI was at the heart of Dr. Prestes’ intervention. He emphasized that this responsibility should always be assumed by humans and not by machines. Thus, the proposed Recommendation should be an ethical guiding compass to rebuild the rule of law in the digital world.

During the expert consultation, Small Island Developing States (SIDS) were identified as a key priority that requires special attention in the Recommendation. Many low- and middle-income countries and most SIDS are labour-intensive economies, and as such can be vulnerable to the effects of automation. Furthermore, reference was made to creating safe and adequate environments for the development of AI, but also the importance of enhancing local capacities including through education so that SIDS and low and middle-income countries, can also develop expertise in relation to the use of AI technologies.

Other concerns raised by the experts revolved around the inclusive dimension of AI, with an emphasis on eliminating inequalities in access, addressing gender bias, and the need to further recognize other populations groups such as vulnerable populations including people with disabilities. The issues of privacy, transparency, data protection, intellectual property rights, regulatory sandboxes, legislation and accountability as well as the role of governments, international organizations and the private sector were also brought up. The use of AI in warfare, as well as its impact on the environment were other topics broached by the experts.

This Virtual Consultation for the Caribbean sub-region is part of a series of multi-stakeholder consultations to be held around the world in August 2020. The consultation process is based on three components including a public online consultation, the Regional and Sub-Regional virtual consultations co-organized with host countries/institutions in all regions and Open, multi-stakeholder, and citizen deliberation workshops in 25 countries around the world .The diversity of perspectives collected during the consultations will be taken into account by the AHEG when revising the first version of the draft text in September 2020.