Fulfilling the economic, social and cultural rights of people of African descent in the age of digitalization, artificial intelligence and new and emerging technologies

The present report is being submitted to the Human Rights Council in accordance with Council resolution 54/26. In the report, the Working Group of Experts on People of African Descent provides brief summaries of the organization and proceedings of its thirty-third and thirty-fourth sessions. The thematic section of the report is based on the Working Group’s consultations with various stakeholders and partners and on research undertaken by the Working Group.

In preparation for the present report, the Working Group issued a call for written submissions from States and civil society. It also held a virtual consultation on 23 July 2024 with experts specialized in various areas of digitalization, artificial intelligence and new and emerging technologies.

In the report, the Working Group stresses that digitalization, artificial intelligence and new and emerging technologies, while inherently neutral and objective, are subject to the perceptions, perspectives, postures and positions of the people who design, develop and deploy those technologies. The regulatory regimes for and the preparation of and for these technologies should therefore include the participation and expertise of people of African descent more fully and ensure that, in addition to having access, they are able to contribute to the shaping of the discourse around and interrogation of these technologies. In that regard, the Working Group recommends measures that would entail a reimagining of data collection, analysis and data use with a view to ensuring appropriate safeguards and the realization of economic, social and cultural rights for people of African descent.

The Working Group thanks States and representatives of international organizations, academic institutions and civil society for their inputs to the report.

The Working Group held its thirty-third session in Addis Ababa, from 4 to 8 December 2023, in closed meetings, and its thirty-fourth session in Geneva, from 22 to 26 April 2024, also in closed meetings. The April session was held in private pursuant to a decision taken by the Working Group to invert, beginning in 2024, its private and public sessions to align the private session with the scheduling of the annual session of the Permanent Forum on People of African Descent for enhanced collaboration and complementarity.

The Working Group held its thirty-third session in Addis Ababa, from 4 to 8 December 2023, in private. This was the first time that a Working Group session had been held outside Geneva and New York. The Working Group chose Addis Ababa mainly to continue to strengthen its collaboration and cooperation with the African Union on two of the Working Group’s current priorities, namely racial justice and reparatory justice, building on prior engagement with the African Commission on Human and Peoples’ Rights and the Citizens and Diaspora Directorate of the African Union, and to engage with other key stakeholders in the region and benefit from a wide participation of people of African descent and African civil society. In addition to its internal planning, monitoring and evaluation agenda, the Working Group met with the African Commission’s outgoing Commissioner and Special Rapporteur on Refugees, Asylum Seekers, Migrants and Internally Displaced Persons in Africa, and discussed with her common concerns about the situations of African migrants in North America, and in the Middle East and North Africa, including the situation of those seeking to travel to Europe. In that context, the Working Group also held a virtual consultation with the Regional Network for Development and Anti-Racism in the Middle East and North Africa and an in-person meeting with a consortium of civil society organizations to inform them about the mandate of the Working Group, exchange ideas and strategies about common priorities and explore opportunities for collaboration to support anti-racism work in the region. The Working Group also held bilateral meetings with the Deputy Regional Representative of the Regional Office for East Africa of the Office of the United Nations High Commissioner for Human Rights; the Special Representative of the Secretary-General to the African Union; the Principal Political Affairs Officer, Office of the Special Envoy of the Secretary-General for the Horn of Africa; the Commissioner of the African Commission on Human and Peoples’ Rights; and the Head of the African Union Citizens and Diaspora Directorate. The Chair of the Working Group, Barbara G. Reynolds, had a virtual meeting with the Deputy Secretary-General on 8 December 2023. Also in the context of the session, the Working Group jointly organized and hosted, with a professor at the University of the Western Cape, John-Mark Iyi, and a professor at Addis Ababa University, Mesenbet Assefa, a colloquium on migration, in a hybrid format, anchored from the Ras Mekonnen Hall at Addis Ababa University, with the participation of the School of Law, Addis Ababa University; the Faculty of Law, University of the Western Cape; the African Centre for Transnational and Criminal Justice; and the Africa Reparations Hub of the University of the Free State, South Africa. The colloquium included presentations from representatives of several universities and from the Senior Migration Advisor of the African Union, which served as inputs for the Working Group’s ongoing research on migration issues affecting people of African descent.

The Working Group held its thirty-fourth session in Geneva, from 22 to 26 April 2024, in private. The session included meetings with United Nations entities working on common areas of focus, and segments on the elaboration of the Working Group’s strategic engagement, advocacy and partnership at the regional level. The Working Group met with representatives of the World Health Organization and the International Labour Organization and held an informal consultation with a delegation of Australia. The Working Group reviewed draft regional strategies following the designation of a focal point for each strategy at its thirty-third session and agreed on approaches for implementation. It also reviewed the draft outline for the present report, and the concept note and workplan for the thirty-fifth session, which will be focused on the theme of reparatory justice and people of African descent. The Working Group reviewed methodologies, lessons learned from the previous three country visits and its working methods, with a view to finalizing agreed changes at its thirty-sixth session.

In July 2023, the Working Group of Experts on People of African Descent conducted virtual consultations with civil society organizations and national human rights institutions and with former members of the Working Group. The consultations informed the preparation of its report entitled “Facilitating the journey from rhetoric to reality”,¹ which was focused on the 20 years of work undertaken since the establishment of the Working Group, the status of implementation of the Working Group’s country and thematic recommendations, its achievements and recommendations for the future. The Working Group also received information from States in response to a call for inputs for that report.

1. A/HRC/54/71. ↩

In September 2023, during the fifty-fourth session of the Human Rights Council, the Chair of the Working Group presented the reports on the Working Group’s country visit to Australia, conducted from 12 to 20 December² 2022,2 and on its country visit to the United Kingdom of Great Britain and Northern Ireland, conducted from 18 to 27 January³ 2023.3 The Chair also presented the consolidated annual report of the Working Group, focused on the economic empowerment of people of African descent,⁴ summarizing the conclusions and recommendations of the Working Group’s thirty-second session, which had been focused on 1 A/HRC/54/71. 2 A/HRC/54/67/Add.2. 3 A/HRC/54/67/Add.1. 4 A/HRC/54/67 and A/HRC/54/67/Corr.1. GE.24-17448 the same theme, and the aforementioned report entitled “Facilitating the journey from rhetoric to reality”.

2. A/HRC/54/67/Add.2. ↩
3. A/HRC/54/67/Add.1. ↩
4. A/HRC/54/67 and A/HRC/54/67/Corr.1. ↩

On 30 October 2023, during the seventy-eighth session of the General Assembly, the Chair of the Working Group presented the above-mentioned consolidated annual report.⁵ She also participated in an interactive dialogue.

5. See A/78/277. ↩

The Chair of the Working Group participated in the twenty-first session of the Intergovernmental Working Group on the Effective Implementation of the Durban Declaration and Programme of Action, held from 16 to 20 October and 20 to 24 November 2023. She briefed participants on the situation of people of African descent in the context of progress achieved in the implementation of the International Decade for People of African Descent. Furthermore, through the participation of the Chair, the Working Group continued its engagement in the drafting of a United Nations declaration on the respect, protection and fulfilment of the human rights of people of African descent. The Chair also participated in the twenty-second session of the Intergovernmental Working Group, held in Geneva from 20 to 24 May 2024, where she presented additional contributions of the Working Group with respect to the process of drafting the declaration.

The Working Group undertook fact-finding country visits to Norway in December 2023 and to Colombia in May 2024. The report on the visit to Norway⁶ will be presented to the Human Rights Council in 2024 and that on Colombia in 2025.

6. See A/HRC/57/70/Add.1. ↩

On 15 and 16 January 2024, the Chair of the Working Group and one other member of the Working Group, Bina D’Costa, participated as panellists at the regional meeting for the Asia and Pacific region on the International Decade for People of African Descent. During the panel focused on recognition, the Chair of the Working Group delivered a statement in which she expressed concern about the invisibility and limited voice and agency of people of African descent in the Asia and Pacific region. During the panel focused on justice, Ms. D’Costa noted that, to date, none of the countries of Asia and the Pacific had invited the Working Group for a country visit or responded to a call for information sent by the Working Group, and that that was reflective of the prevalence of racial discrimination, manifested in a denial of the existence of people of African descent in the region.

On 11 April 2024, the Chair participated in meetings focused on cooperation and an exchange of views with the European Commission against Racism and Intolerance and other European stakeholders, organized by the European Commission and held in Strasbourg, France.

The Working Group continued to issue communications to States concerned regarding cases of racial discrimination faced by people of African descent.

Among other activities, the Chair: (a) Delivered a keynote address for a webinar on co-creating solutions for people of African Descent in the context of recognition, justice and development, organized by the African Australian Advocacy Centre (11 August 2023); (b) Served as a panellist in a virtual meeting on first-person testimonies of people of African descent fleeing Ukraine, organized by the Working Group and People of African Descent Link (31 August 2023); (c) Recorded a video message, issued by Office of the United Nations High Commissioner for Human Rights, for the commemoration of the seventy-fifth anniversary of the Universal Declaration of Human Rights (10 December 2023); (d) Delivered a video presentation during a side event at the fifty-fifth session of the Human Rights Council, entitled “Combating global racism: implementing the Durban Declaration and Programme of Action”, organized by a group of non-governmental organizations (25 March 2024); 5 See A/78/277. 6 See A/HRC/57/70/Add.1. GE.24-17448 (e) Served as a panellist in a side event at the third session of the Permanent Forum on People of African Descent, on implementing the Durban Declaration and Programme of Action as an integral part of the International Decade for People of African Descent, organized by a group of non-governmental organizations (16 April 2024); (f) Served as a panellist in a virtual meeting focused on the contribution of the right to development towards combating discrimination, including racial discrimination, organized by the Working Group on the Right to Development (15 May 2024); (g) Served as a panellist in a virtual meeting on Summit of the Future dialogues, focused on the economic empowerment of people of African descent in the context of financial architecture, governance and the global monetary and financial systems, organized by Geledés – Instituto da Mulher Negra (7 June 2024); (h) Served as a panellist at a conference entitled “Securing Afrodescendent peoples’ land tenure rights in Latin America and the Caribbean: an effective pathway to conservation and climate change action”, organized by Rights and Resources Institute, Inc. and others and held in Bogotá from 11 to 14 June 2024; (i) Served as a panellist in a virtual meeting, during a side event at the 2024 high-level political forum on sustainable development, on strengthening the role of major groups and other stakeholders in combating global racism (17 July 2024).

Activities undertaken by other members of the Working Group included the following:
(a) Ms. D’Costa:
(i) Organized and chaired a meeting between the Working Group and the Expert Mechanism on the Rights of Indigenous Peoples, held in Canberra, in which the plans under both mandates for the next five years were discussed (16 October 2023);
(ii) Delivered a presentation on the Working Group’s mandate and work at a conference in Australia (December 2023);
(iii) Gave a presentation on victims and survivors of trafficking at the sixth meeting of the Bali Process Technical Experts Group on Returns and Reintegration, at the invitation of the International Organization for Migration and the Government of Australia (March 2024);
(iv) Served, with former New Zealand Prime Minister Helen Clark and Australian Research Council Laureate Fellow and Professor of Health Equity Sharon Friel, as a panellist at a high-profile event on how to build a healthier future for all, held at the Australian National University, during which she discussed migration and asylum issues and racism against people of African descent, with a focus on the Working Group (April 2024);
(v) Presented, with Shine Choi, the Working Group’s work at a conference convened by the International Feminist Journal of Politics and Feminist Africa in Maputo, through a paper on anti-racist education in Asia and the Pacific and the engagement of international human rights mechanisms for the rights of Africans and people of African descent (July 2024);
(vi) Was the keynote speaker at the Development Futures Conference, organized by the Development Studies Association of Australia at the University of Melbourne, where she presented a paper on hopelessness and racism and also spoke about decolonizing the curriculum in Australia and elsewhere in the Asia and Pacific region (July 2024);
(vii) Served as a panellist in a discussion on Indigenous rights at the United Nations, organized by the Coral Bell School of Asia Pacific Affairs, Australian National University (16 July 2024);
(viii) Served as a panellist at a session on the theme “Resistance, power, and the new global ethical order”, held during a symposium to mark the seventy-fifth anniversary of the Department of International Relations at the Australian National University; she spoke about racism, colonialism and the invisibility of people of African descent in the war in the Occupied Palestinian Territory (18 July 2024);
(b) Catherine S. Namakula:
(i) Participated as a moderator at a hybrid colloquium on the theme “Unifying Africa for action to advance reparatory justice”, which had the overall aim of fostering dialogue and action towards reparatory justice in Africa and which was held in the context of the launch of the University of the Free State Africa Reparations Hub (June 2024);
(ii) Continued her work with the African Union reparations desk, where the current focus is on securing funding through dialogue with funders.

Digitalization, artificial intelligence and new and emerging technologies, and their processes, have led to significant transformations across virtually all sectors of society. New and emerging technologies are generally characterized by the following five attributes: radical novelty; relatively fast growth; coherence; prominent impact; and uncertainty and ambiguity.⁷ Those attributes provide the basis for defining a new and emerging technology as:
A radically novel and relatively fast-growing technology characterised by a certain degree of coherence persisting over time and with the potential to exert a considerable impact on the socio-economic domain(s) which is observed in terms of the composition of actors, institutions and patterns of interactions among those, along with the associated knowledge production processes.⁸

7. Daniele Rotolo, Diana Hicks and Ben R. Martin, “What is an emerging technology?”, Research Policy, vol. 44, No. 10 (December 2015); Oleg Litvinski, “Emerging technology: toward a conceptual definition”, International Journal of Trade, Economics and Finance, vol. 9, No. 6 (December 2018); and Winston & Strawn LLP, “Law glossary: what is emerging technology?” (2024), available at https://www.winston.com/en/legal-glossary/emerging-technology. ↩
8. Rotolo, Hicks and Martin, “What is an emerging technology?”, p. 13. ↩

Artificial intelligence should be understood as one among many new and emerging technologies. Digitalization and artificial intelligence have enormous potential for automating and augmenting human decision-making and behaviour. Their impact on the political, civil, economic, social and cultural sectors is rapidly becoming evident, raising substantial questions and concerns about human rights and ethics, as the use, misuse and abuse of those technologies can inadvertently or intentionally introduce, perpetuate and exacerbate biases and discrimination in virtually all sectors, communities and societies. Digitalization and artificial intelligence have a multiplier effect, extending ever more rapidly their scope and reach with each successive iteration in their design and deployment, whether for good or for ill. Their impact is therefore not merely aggregative but also cumulative, with each successive use, adaptation and application of an earlier technology building on what already was in place in multiple ways, faster and further.

While digitalization and artificial intelligence are inherently neutral and objective, they are designed, developed and deployed by human beings. Human beings have perceptions and perspectives, postures and positions on every issue. They are therefore inherently subjective, naturally biased and have the capacity to be either potentially discriminatory and damaging or equalizing and enabling to other human beings. They consciously or unconsciously integrate those capacities into digital technologies, which in turn automate or augment the capacities. The ethical implications and ramifications of digital technologies for human beings and human ecology are therefore a critical area of concern. In his 12 July 2023 address at a high-level side event during the fifty-third session of the Human Rights Council, the United Nations High Commissioner for Human Rights noted that “to be effective, to be humane, to put people at the heart of the development of new technologies, any solution – any regulation – must be grounded in respect for human rights”.⁹ The Committee on Social Affairs, Health and Sustainable Development of the Parliamentary Assembly of the Council of Europe has noted that artificial intelligence “is aiming to replace humans in strenuous mental tasks rather than physical ones”, and that physical tasks had been targeted in previous waves of automation and robotization.¹⁰ For people of African descent who routinely face institutional, structural and systemic racism, whether in the use of everyday apps or as a result of complex algorithms, digitalization and artificial intelligence have the potential to mask, accelerate, intensify, broaden and deepen discrimination while appearing neutral and even benevolent when compared with the non-digital racist practices of previous eras. They also offer unprecedented opportunities to protect and promote human rights, democracy and the rule of law.

9. See https://www.ohchr.org/en/statements/2023/07/artificial-intelligence-must-be-grounded-human-rights-says-high-commissioner. ↩
10. “Artificial intelligence and labour markets: friend or foe?” (2020), citing Ekkeharad Ernst, Rossana Merola and Daniel Samaan, “The economics of artificial intelligence: implications for the future of work” (Geneva, International Labour Organization (ILO), 2018). Available at https://pace.coe.int/en/files/28738/html. ↩

While digital technologies can contribute to solutions to societal problems, their unpredictable nature and uncontrolled use and inexplicability, and the reflection or amplification of data biases, raise various concerns about privacy, security, fairness, human rights and even democracy. Increasing evidence demonstrates how existing and emerging technologies might not only exacerbate existing inequalities, but also differentiate and target marginalized communities. Racialized communities are disproportionately affected in a negative way by surveillance, (data-driven) profiling, discrimination online and other violations of digital rights. While these technologies are increasingly deployed to address complex social problems, their impact on people of African descent is systematically overlooked.

The Working Group held a virtual consultation on 23 July 2024 with technology experts engaged in digitalization, artificial intelligence and new and emerging technologies. The consultation was an opportunity to discuss the racialization, documented misuse and actual and potential discriminatory impact of digitalization, artificial intelligence and new and emerging technologies on the economic, social and cultural rights of people of African descent.

Focusing on African economic independence in the age of artificial intelligence and digitalization, Kofi Takyi Asante, a senior research fellow at the Institute of Statistical, Social and Economic Research at the University of Ghana, highlighted the revolutionary potential of artificial intelligence and digitalization and explained that it involved risks as well as enhanced efficiency and productivity. Citing a 2024 report, he stated that by 2030, artificial intelligence was projected to contribute $15.7 trillion to global gross domestic product and that China and North America would gain the most.¹¹ He further stated that of that $15.7 trillion, $1.2 trillion could be generated in Africa.¹² In Africa, many sectors, including health, agriculture, water and clean energy, could benefit from the application of artificial intelligence. The application of artificial intelligence could contribute to the achievement of Agenda 2063: The Africa We Want, of the African Union, and the Sustainable Development Goals. Artificial intelligence development was currently concentrated in China, the United Kingdom and the United States of America. Some universities in Africa including the University of Pretoria, Makerere University and Kwame Nkrumah University of Science and Technology, had opened artificial intelligence labs, and the African Institute for Mathematical Sciences had developed master’s and/or doctoral programmes in artificial intelligence, machine learning, mathematics and data science. Many of the major multinational technology firms had opened artificial intelligence labs in Africa, including Microsoft in Nairobi in 2020, Google in Accra in 2018, and IBM in Nairobi in 2018 and in Johannesburg in 2016. A growing number of artificial intelligence startups, albeit isolated and small-scale, had been emerging across the continent. In Ghana, artificial intelligence and digitalization initiatives had resulted in the development of the Ghana Card, a multipurpose biometric national identification card for day-to-day electronic and physical transactions; a digital system for addresses; drone delivery of medical products; and artificial intelligence systems to help farmers track weather patterns. Challenges to artificial intelligence development in Africa included low investment in research and development and a lack of relevant skills; there was a need for supportive policies and robust infrastructure to enable Africa to benefit fully from artificial intelligence.¹³ Most African countries lacked the financial, technological and institutional capacity to drive artificial intelligence development, owing to how the continent’s development had been undermined by foreign imperialist interests, including through the international financial architecture, which contributed to disinvestment from social sectors such as education.

11. PwC, “Sizing the prize: what’s the real value of AI for your business and how can you capitalise?” (2017). According to figures in the report, China is projected to experience gains amounting to 26 per cent of its gross domestic product; in North America, the gains are projected to be 14.5 per cent of gross domestic product. ↩
12. See https://www.un.org/africarenewal/magazine/march-2024/artificial-intelligence-and-africa. ↩
13. Ibid. ↩

Joe Atkinson, of the University of Southampton, focused on human rights at work in the age of artificial intelligence. He stated that governance informed by artificial intelligence and algorithmic decision-making had emerged as a new form of “governance by numbers” in both the public and the private sectors. Algorithmic decision-making was being used by governments for a wide range of decisions relating to policing, immigration, housing and social security. It was also being used by corporations for targeted advertising and recommendations and for personal pricing. Mr. Atkinson detailed the use of artificial intelligence in automation and algorithmic management in areas such as recruitment, including for pre-screening, curriculum vitae sifting and interview analysis; in route planning and the scheduling of allocation of resources (for example in platforms or apps); in evaluation, to monitor tasks and performance through algorithmic ratings and assessments (for example at call centres); in discipline, in order to, inter alia, suspend low-scoring workers and alter access to shifts; and in dismissals and redundancies, which could be affected by a reliance on algorithmic metrics. Automation could eventually lead to a level of job destruction and work scarcity that undermined the right to work, in which case policies designed to protect the right to work would be needed. Such protections could entail limitations on the automation of specific tasks or jobs; policies that spread work across more people; and job guarantee schemes. Technology also threatened equality at work: algorithmic management posed a serious threat to the right to non-discrimination. The threat could stem from assumptions or bias of engineers; inaccurate or incomplete data, leading to errors or biases; and the replication or amplification of existing inequalities. The problem was compounded by a lack of transparency of and accountability for the design and implementation of algorithmic tools. Technology also posed a threat to the right to just and fair working conditions in numerous ways. Algorithmic management undermined just conditions by enabling the avoidance of employment law protections. Surveillance and intensification of work created health and safety risks. It heightened the level of control over and subordination of workers. It also enabled the deskilling of work; the overall effect was to recommodify and dehumanize work.

Isak Nti Asare, a professor at the Center for Applied Cybersecurity Research at Indiana University, drawing from Johan Galtung’s 1969 paper entitled “Violence, peace and peace research”,¹⁴ focused on the application of positive peace in building governance mechanisms for artificial intelligence and emerging technologies. He underscored the understanding that technology was a product of the underlying society in which it was conceived, which included attitudes, structures and institutions. He warned that inequalities in contemporary technological tools and systems were predicated on the consolidation of power in the digital economy among a few technology companies, and were only to be expected given such a primordial environment of structural inequality. The current global focus on a paradigm of harm mitigation within the digital ecosystem addressed the symptoms of an otherwise structural problem, which consequently fuelled the manifestation of inequality and violent outcomes.

14. Journal of Peace Research, vol. 6, No. 3 (1969). ↩

Gift Mwonzora, an expert from the Willy Brandt School of Public Policy at the University of Erfurt, focused on health and healthcare, nutrition and food security. He presented findings from his research and highlighted examples of precarity in the labour market as a result of automation and digitalization in agriculture. In South Africa, a large proportion of women and girls, in particular women of colour, in the labour force were employed in citrus farming. While picking fruit largely remained a physical activity still involving human intervention, mechanization in some processes of production, such as in sorting and grading, had resulted in women fearing the loss of, or indeed losing, their jobs. That had further compounded existing vulnerabilities related to the casualization of labour, low wages and the seasonal nature of work in agriculture. Increasingly, women workers in such sectors were being replaced by artificial intelligence-driven production processes. In another sector, the use of drone delivery systems in the medical field in Malawi and Rwanda was addressing some of the challenges of lack of access to healthcare in remote areas, constituting a positive use of artificial intelligence and technology. However, ethical concerns remained owing to a lack of duty of care regarding people of African descent in medical trials.

The technology sector has been criticized for its lack of diversity, favouring white, affluent men. Large-scale artificial intelligence systems are developed almost exclusively in a small number of companies and elite university laboratories that employ mostly white men and have a history of discrimination against and exclusion of “others”, including people of African descent. Technology that is developed and produced in environments that disproportionately exclude people of African descent are more likely to reproduce racial inequalities.

The creation of artificial intelligence systems begins with data, namely their extraction and organization and subsequent modelling. Each step in this process holds the potential to introduce or perpetuate racial bias, significantly affecting healthcare outcomes for people from certain racial or ethnic groups. Artificial intelligence systems are trained on enormous quantities of data, mostly on non-Black populations, which are used to build models of behaviour. The designers and developers of machine learning and artificial intelligence systems can therefore intentionally or unintentionally introduce biases into their algorithms through the use of prebuilt models that contain racial biases, as evident in some generative artificial intelligence systems being unable to create accurate and realistic depictions of Black people. How developers obtain such critical data raises ethical questions. Data acquisition practices often lack transparency, with instances where data are obtained without proper consent or through exploitative means.

Face recognition software used by governments and the police disproportionately affects people of African descent. Such software learns and propagates biased associations between race groups and negative attributes, exacerbating racial inequality. In 2015, for example, Google had to apologize after its image-recognition app mistakenly labelled African Americans as “gorillas”.

The surveillance practices used in times of enslavement and colonization, and that still persist, can and have been made worse with the use of artificial intelligence. Research has consistently shown greater inaccuracies in the application of those practices to members of non-white populations. This has already led to several dangerous situations for people of African descent, such as being falsely identified as a suspect for a crime. Accounts of the disproportionate levels of harm from face recognition software experienced by people of African descent are well known.

The lack of transparency and accountability in artificial intelligence development exacerbates these issues. Many artificial intelligence systems are developed and deployed by private companies that do not disclose their algorithms’ inner workings, citing proprietary concerns. This opacity makes it difficult for independent researchers, policymakers and the communities affected to scrutinize and challenge biased algorithms. Without transparency, it is nearly impossible to hold developers accountable for the adverse impacts of their technologies on marginalized groups. Moreover, often there are no mechanisms in place to audit or regulate artificial intelligence systems effectively. Regulatory bodies lack the technical expertise and resources needed to assess the fairness and accuracy of complex algorithms. This regulatory gap allows biased artificial intelligence systems to proliferate unchecked, further entrenching existing social and economic disparities.¹⁵

15. Submission by Motse Ntloedibe-Kuswani, of the American University of Paris. ↩

Although racial bias and discrimination in housing has been well documented for decades, it is not yet clear how digitalization and artificial intelligence are affecting access to housing and related basic services such as water, sanitation and electricity. For example, in the United States, racial bias in housing has manifested in the lower valuation of housing in predominantly Black neighbourhoods by around 21 to 23 per cent as compared with housing in non-Black neighbourhoods. Homes in majority-Black neighbourhoods are 1.9 times more likely to be appraised at below the contract price as compared to homes in majority-white neighbourhoods.¹⁶ This pattern of residential segregation can also be observed in other countries, such as Brazil.

16. Jonathan Rothwell and Andre M. Perry, “How racial bias in appraisals affects the devaluation of homes in majority-Black neighborhoods” (Brookings Institution, 5 December 2022). ↩

Artificial intelligence is revolutionizing healthcare, offering the potential to enhance diagnostic accuracy, streamline patient care and improve health outcomes. Initially conceived to revolutionize clinical decision-making and patient care, artificial intelligence is heavily reliant on vast data sets comprising diverse sources: patient histories, genetic profiles, lifestyle data and more. If training data predominantly represent majority groups, the resulting models are inherently biased, producing recommendations and predictions that favour those populations. Also, those data are often extracted and used without robust oversight or clear consent frameworks, raising significant concerns about fairness and equity. Such practices can inadvertently embed biases into artificial intelligence models, perpetuating disparities in healthcare outcomes, affecting people of African descent in particular. The ethical principles guiding this extraction and implementation are often left to the discretion of the developers owing to a significant lack of regulation. This regulatory gap means that embedding ethical considerations from the outset is crucial in order to ensure that artificial intelligence technologies benefit patients rather than cause harm. The bias is compounded by decisions in feature engineering and hyperparameter tuning, which may overlook factors critical to understanding and addressing health disparities among people of African descent.

The opacity of these algorithms – often referred to as the “black box problem” – obscures how decisions are made, making it challenging to identify and rectify biases that disadvantage patients of African descent. Research indicates that health data are predominantly skewed towards white and male populations, reflecting historical social practices and individual programmer biases that shape artificial intelligence systems. An important study scrutinized a widely adopted artificial intelligence algorithm in United States healthcare, revealing a troubling bias favouring white patients over equally ill Black patients. The algorithm’s reliance on historical cost data disadvantaged patients of African descent owing to lower previous healthcare expenditures that were influenced by socioeconomic factors. Consequently, Black patients received fewer critical medical interventions, exacerbating health disparities and revealing systemic biases against people of African descent within artificial intelligence systems. The algorithm relied on healthcare spending to predict future health needs. But with less access to care historically, patients of African descent often spent less. As a result, they had to be much sicker to be recommended for extra care under the algorithm.¹⁷

17. Ziad Obermeyer and others, “Dissecting racial bias in an algorithm used to manage the health of populations”, Science, vol. 366, No. 6464 (25 October 2019). ↩

The coronavirus disease (COVID-19) pandemic was a reminder of both the promise of artificial intelligence and the urgent need to strike a balance between the collective interest and individual rights. The crisis brought to light issues regarding data access and sharing, liability, data and algorithm quality, the complementarity of technology and the human, and the need for interdisciplinary cooperation and collaboration.¹⁸

18. Committee on Social Affairs, Health and Sustainable Development of the Parliamentary Assembly of the Council of Europe, “Artificial intelligence in health care: medical, legal and ethical challenges ahead”, available at https://pace.coe.int/en/files/28737/html. ↩

Data cleaning is undertaken to refine data sets by removing anomalies and normalizing information. Yet, what constitutes an anomaly may differ significantly across demographic groups. Data cleaning processes that overlook common occurrences within diverse populations of African descent further marginalize the health concerns of those groups, leading to incomplete and biased data sets that fail to capture the full spectrum of health issues faced by people of African descent.

In 2022, the European Union established the European Health Data Space to create a common framework for the secondary use of health data. However, ethical concerns persist, particularly regarding patient consent practices. The lack of informed consent exacerbates issues of privacy and data security. Furthermore, there are risks of inequitable access to data, favouring larger, well-funded organizations over smaller entities.

The deployment of artificial intelligence in agriculture often undermines food sovereignty by prioritizing technological solutions over traditional knowledge and practices. Digital technologies and artificial intelligence can displace small-scale farmers, erode peasant knowledge and concentrate power in the hands of large corporations.¹⁹ This shift threatens the autonomy of local communities and their ability to sustainably manage their food systems. Developing artificial intelligence technologies that incorporate local knowledge and building local capacities to own and manage these systems can reduce dependency on external actors.

19. Friends of the Earth Europe, FIAN International and the Centre for Agroecology, Water and Resilience at Coventry University, “Remote control and peasant intelligence: on automating decisions, suppressing knowledges and transforming ways of knowing” (2023). ↩

Given that to date the private sector has driven most of the research and development of artificial intelligence applications for healthcare, national public healthcare authorities should adopt a strategic approach to coordinating digitalization policies, research and investment, and to the management and use of personal data, with a view to ensuring full protection of fundamental rights and striking a healthy balance among individual, business and public interests.²⁰

20. Parliamentary Assembly of the Council of Europe, recommendation 2185 (2020), entitled “Artificial intelligence in health care: medical, legal and ethical challenges ahead”. ↩

Education and training, employment and economic empowerment are, in the context of digitalization, both means and end. Education and training, alongside the knowledge, skills, expertise and attitudes that they produce, are in turn foundational to employment and, alongside employment, foundational to economic empowerment. Literacy, including digital literacy, drives development, enables participation in the labour market, reduces poverty and enhances individual and collective performance on most social development indicators.²¹ At the moment, the technological divide is clear, although inconsistently so across some platforms,²² leaving people of African descent far behind others in digital literacy, fluency and competence because of resource constraints (availability and affordability of devices, Internet access and access to electricity).

21. United Nations Educational, Scientific and Cultural Organization, “Report of the Global Conference on promoting literacy for a world in transition: building the foundation for sustainable and peaceful societies” (Paris, 2023). ↩
22. Aaron Smith, “African Americans and technology use: a demographic portrait”, Pew Research Center (6 January 2014); and Community Tech Network, “Digital equity for Black Americans: a racial justice issue” (6 February 2023). ↩

There is a strong correlation between poverty and the level, quality and quantity of education, and correspondingly between racialized poverty and racialized education. The digital divide in education is generally aligned with this existing divide. In poor neighbourhoods and communities, there is less access to digital education from the preschool level to post-secondary levels because of limited access to electricity or alternative energy sources, to digital devices, to digital content and, ultimately, to teachers and university professors who themselves are digitally competent. Where people of African descent are poorer than other groups, the intersection between poverty and race has resulted in digitally disadvantaged homes, schools and communities. This means that children of African descent enter higher levels of education and training disadvantaged and consequently are less prepared to work in the digital world.

At higher levels of education, race plays a triple role in digitalization and the use of artificial intelligence. First, there is much less research about the issues and concerns of people of African descent, including – and particularly – by scholars and researchers of African descent. Consequently, the frameworks used for such research may, intentionally or unintentionally, introduce and perpetuate racial bias and prejudice, rendering the data they produce injurious to people of African descent. Second, artificial intelligence is predicated on the use of large data sets; the paucity of accurate data about people of African descent presents another layer of bias. Third, biases and stereotypes are deeply embedded in machine learning,²³ as for example in the field of photography, where “photographic systems attempted to create a universal or neutral standard for all subjects, yet the norm ended up being white skin”.²⁴

23. Ludovica Marinucci, Claudia Mazzuca and Aldo Gangemi, “Exposing implicit biases and stereotypes in human and artificial intelligence: state of the art and challenges with a focus on gender”, AI & Society, vol. 38, No. 2 (2023); and Ryan S. Baker and Aaron Hawn, “Algorithmic bias in education”, International Journal of Artificial Intelligence in Education, vol. 32, No. 4 (2022). ↩
24. Nettrice R. Gaskins, “Interrogating AI bias through digital art”, Just Tech, Social Science Research Council (7 September 2022). ↩

In its 2021 flagship report World Employment and Social Outlook: The Role of Digital Labour Platforms in Transforming the World of Work, International Labour Organization (ILO) documented the pervasive nature of digitalization and artificial intelligence in virtually all aspects of work in diverse sectors, such as agriculture, transport and industry, and the ubiquitous nature of all types of learning platforms. Researchers have found evidence that concerns about exposure to the effects of digitalization and artificial intelligence and potential job losses are justified in some areas. Routine jobs, depending on the overall development, wealth and technological and industrial advancement of the society, are particularly vulnerable to automation. The researchers also indicated that machine learning systems were also able to improve performance in non-routine tasks. They posited that in low-income countries, only 0.4 per cent of total employment was potentially exposed to automation effects, versus 5.5 per cent of total employment in high-income countries. The impact of augmentation is higher: 10.4 per cent of employment in low-income countries, and 13.4 per cent in high-income countries, is affected.²⁵ This has an impact on people of African descent.

25. Paweł Gmyrek, Janine Berg and David Bescond, “Generative AI and jobs: a global analysis of potential effects on job quantity and quality”, ILO Working Paper No. 96 (Geneva, ILO, 2023). ↩

The ILO Committee of Experts on the Application of Conventions and Recommendations, in its general observation on discrimination based on race, colour and national extraction, adopted in 2018, notes that under the Discrimination (Employment and Occupation) Convention 1958 (No. 111), the term “race” includes any discrimination against linguistic communities or minority groups whose identity is based on religious or cultural characteristics or national or ethnic origin, with “colour” being one of the ethnic characteristics. The Committee acknowledges the persisting patterns of discrimination on the grounds of race, colour and national extraction.

Researchers have noted that the impact of digitalization and artificial intelligence on employment is not an apocalypse but a shift, arguing that the potential augmentation effects are greater than automation exposure in most countries, and therefore more likely to transform structures and roles than replace them, although some jobs may be lost.²⁶

26. See, for example, Gmyrek, Berg and Bescond, “Generative AI and jobs”. ↩

Racialization of and racial discrimination in employment, carrying over the positive or negative results of racialized education and training, begin in recruitment, and permeate decisions about orientation, professional development and training, job assignments, recognition and reward, and ultimately promotion and separation, whether voluntary or involuntary. Compounding the situation is that the racism in housing, healthcare, justice and other spheres, also fed by racialized generative artificial intelligence, seeps into education and employment. Thus, as digitalization becomes increasingly widespread in the world of work, any biases of employers, including not only those based on race, but also those based on gender and other demographic factors, influence decisions. As it becomes more widely used to analyse and interpret data, artificial intelligence, because of undetected bias, introduces into employment stereotypes and prejudice that work against people of African descent.

Few people of African descent inherit wealth, and intergenerational and transgenerational poverty and the consequential disadvantage are well documented. The carry-over from education through to employment can be observed in levels of economic empowerment among people of African descent. Bias in coding and in algorithms used in the housing, banking, financial and insurance sectors is also well documented. In its 2023 report on the economic empowerment of people of African descent, the Working Group dealt extensively with the bias, prejudice and discrimination against people of African descent in the economic, financial and digital sectors.

Digital inequities acutely affect young people globally, facilitate the spread of disinformation and misinformation and restrict economic opportunities and successes for people of African descent.²⁷ People of African descent are not well represented in data sets, which has an impact on algorithmic decision-making systems, causing disproportionate harm to and resulting in discrimination against people of African descent.

27. See A/HRC/54/67 and A/HRC/54/67/Corr.1. ↩

The result of the situations in education, employment and economic empowerment is graphically played out in the data on patents as an indicator of innovation, research and development. World Intellectual Property Organization data suggest a bleak picture. In 2023, the data showed that computer technology topped patent registration across the globe, with digital communication in the third spot. They also showed that Asia outstripped every other region. With the exception of South Africa, patent growth in countries with large concentrations of people of African descent was negligible.²⁸ While the United States remains the top producer of patents, historical data show that the percentage of patents held by Black Americans (2.7 per cent between 2010 and 2012,²⁹ and 1.1 per cent in 2021³⁰) falls well below the percentage of Black Americans in the overall population (14.4 per cent).³¹ It is important to note that people of African descent, while often concentrated in resource-rich areas, remain socially and economically disadvantaged. This affects their rates of participation in digitalization and in science and technology and thus the advancement of their communities and societies.

28. World Intellectual Property Organization, WIPO “IP facts and figures 2023” (Geneva, 2023). ↩
29. Jonathan Rothwell, Andre M. Perry and Mike Andrews, “The Black innovators who elevated the United States: reassessing the golden age of invention” (The Brookings Institution, 23 November 2020). ↩
30. Zippia, “Inventor demographics and statistics in the US”, 5 April 2024, available at https://www.zippia.com/inventor-jobs/demographics/. ↩
31. See https://www.pewresearch.org/social-trends/fact-sheet/facts-about-the-us-black-population/. ↩

The enjoyment by people of African descent of their cultural rights is also at risk in the digital age. Digital platforms and artificial intelligence algorithms that prioritize certain types of content over others can marginalize cultural expressions of and contributions from communities of African descent. This cultural marginalization not only reduces the diversity of digital spaces but also contributes to the erasure of Black cultural identities and narratives. The lack of representation in the development of digital technologies means that the cultural nuances and needs of people of African descent are often overlooked. This exclusion is particularly evident in content recommendation algorithms on social media and streaming platforms, which may not promote content created by or relevant to Black communities. It is imperative to implement policies and practices that promote meaningful inclusivity and representation in the technology industry to ensure that digital technologies serve the diverse cultural needs of all communities.

Digitalization has the potential to bring diverse people together to build strong relationships through various forms of artistic expression. A growing body of research points towards the value of arts- and sports-based community development initiatives and standpoint projects focused on transformative youth development that are organized in collaboration with people of African descent and that are aimed at supporting people in the exploration and self-expression of their identities and the development of their sense of agency. The value of the arts in renegotiating identities is recognized among people of African descent. Researchers have noted that music can provide “a powerful means of representing identity, and of asserting cultural difference”.³² For example, some researchers have looked at how young members of the African diaspora in Australia create and use the art form of hip-hop to develop and broadcast their political voice in order to express resistance to racialized discourses, to sustain agency, to counter stereotypes and to express self-determination. However, a knowledge gap persists, and digitalization is not always used to benefit people of African descent. Unlimited access to the intellectual property of others remains a major concern for artists of African descent, whose ideas and culture can be appropriated by others in digital spaces and exploited for profit.

32. Liesbeth de Block and David Buckingham, Global Children, Global Media: Migration, Media and Childhood (Palgrave Macmillan, 2007). ↩

The Working Group notes the diversity of measures and initiatives being implemented at the national, regional and international levels to address the moral and ethical concerns about artificial intelligence.

In October 2022, the Food and Drug Administration of the United States issued guidance significantly broadening the scope of the tools it planned to regulate. This guidance reflects a growing acknowledgement that more must be done to combat bias and promote equity amid the growing number and increasing use of artificial intelligence and algorithmic tools.³³ On 30 October 2023, the President of the United States signed the Executive Order on the Safe, Secure, and Trustworthy Development and Use of Artificial Intelligence, which laid out principles for executive departments and agencies, stressing the need for artificial intelligence to comply with all federal laws and for artificial intelligence policies to be consistent with the advancement of “equity and civil rights”. In the Executive Order, the President expressed concern about how artificial intelligence had been used to deepen discrimination and bias. He stated that artificial intelligence systems deployed irresponsibly had “reproduced and intensified existing inequities, caused new types of harmful discrimination, and exacerbated online and physical harms”.

33. Food and Drug Administration, “Clinical decision support software: guidance for industry and Food and Drug Administration staff”, 28 September 2022, available at https://www.fda.gov/media/109618/download. ↩

Brazil reported that in 2023, on 20 November, the National Day of Zumbi and Black Consciousness, it had adopted a decree establishing an interministerial working group to draft a national plan to combat racism in communications.³⁴ The interministerial working group aimed to propose measures to promote rights and combat racism in digital communication services, and to strengthen and sustain Black media. In the framework of the interministerial working group, the Ministry of Racial Equality and the Social Communication Secretariat had made available an online library of resources in the area of digital technologies and racial justice.³⁵

34. See https://www.planalto.gov.br/ccivil_03/_ato2023-2026/2023/decreto/d11787.htm (in Portuguese). ↩
35. See https://www.gov.br/igualdaderacial/pt-br/assuntos/gti-comunicacao-antirracista/biblioteca (in Portuguese). ↩

In Colombia, the Ministry of Cultures, Arts and Knowledge has initiated the consideration of protocols for the ethical and democratic use of artificial intelligence within the Ministry, with a view to guaranteeing the sociocultural rights of people of African descent and other ethnic groups. In the area of education, the Ministry and the National University of Rosario are working together to better understand, and to raise awareness of, the issue of culture and artificial intelligence.

Ecuador underlined the impact of artificial intelligence, digitalization and new and emerging technologies on the fundamental and collective rights of people of African descent, and the potential risk posed by these tools to the socioeconomic rights of people of African descent. In June 2024, the National Assembly had registered the presentation of a draft organic law for the regulation and promotion of artificial intelligence in Ecuador. Ecuador had urged the legislature, and public and oversight institutions, to draft rules and regulations for the development, provision, commercialization and utilization of artificial intelligence, digitalization and new and emerging technologies based on human rights, including the rights of peoples, members of Indigenous nations and people of African descent.

Spain reported that its Law No. 15/2022 addressed equal treatment and non-discrimination in the field of artificial intelligence and automated decision-making. Furthermore, the country’s Strategic Framework for Citizenship and Inclusion, against Racism and Xenophobia (2023–2027) guaranteed equal opportunities in access to information and communications technology.

On 22 September 2021, the United Kingdom published its 10-year National AI Strategy.³⁶ The Government has since engaged in a public consultation with regulators on the White Paper of March 2023 on artificial intelligence regulation, in which the Government proposed a framework of five voluntary principles to guide responsible artificial intelligence design, development and use. On 1 and 2 November 2023, the Government hosted the AI Safety Summit, culminating in the Bletchley Declaration, an agreement among 28 countries and the European Union on cooperation and responsibility-sharing to mitigate the risks of artificial intelligence.

36. See https://assets.publishing.service.gov.uk/media/614db4d1e90e077a2cbdf3c4/National_AI_Strategy_-_PDF_version.pdf. ↩

The African Union aims to harness the revolutionary potential of artificial intelligence in the fulfilment of the long-term development aspirations set out in its Agenda 2063, while recognizing the risks of artificial intelligence and the need for artificial intelligence development on the continent to be Africa-owned, people-centred and inclusive and for it to meet the specific needs of Africans, including in education, healthcare, agriculture, infrastructure, peace and security and good governance. To that end, on 13 June 2024, ministers responsible for information and communications technology endorsed the Continental Artificial Intelligence Strategy and the African Digital Compact³⁷ at the second extraordinary session of the Specialized Technical Committee on Communication and Information Communications Technology of the African Union.

37. See https://au.int/sites/default/files/pressreleases/43871-pr-Ministerial_PR_-_STC-CICT1.pdf. ↩

The Council of Europe Framework Convention on Artificial Intelligence and Human Rights, Democracy and the Rule of Law is an international legally binding instrument. It was adopted on 17 May 2024 by the Committee of Ministers of the Council of Europe at its 133rd session. The purpose of the Convention is to ensure that artificial intelligence systems, throughout their life cycles, are consistent with human rights, democracy and the rule of law. It sets out principles for States to follow to ensure human dignity and individual autonomy; transparency and oversight; accountability and responsibility; equality and non-discrimination; privacy and personal data protection; reliability; and safe innovation.

The European Union’s regulatory framework for artificial intelligence, known as the European Artificial Intelligence Act,³⁸ entered into force on 1 August 2024. The Act harmonized rules on responsible development and use of artificial intelligence within the European Union, which include transparency obligations in the area of general-purpose artificial intelligence and mandatory labelling of artificial or manipulated images and audio or video content, and a prohibition of the use of biometric identification systems by law enforcement. The purpose of the Act is to ensure safety and compliance with fundamental rights and democracy while promoting innovation.

38. Regulation (EU) 2024/1689 of the European Parliament and of the Council of the European Union of 13 June 2024. ↩

Digitalization and artificial intelligence, including generative artificial intelligence, are fundamentally neutral. Bias and consequently prejudice, discrimination and the violation of accepted ethical and human rights norms and standards seep into digitalization and artificial intelligence through human use, misuse and abuse. As digitalization and artificial intelligence have a cumulative, compounding and multiplier effect, they amplify, deepen, widen and enmesh their impact, positive or negative, in ever more complex ways. Research shows that people can unconsciously retain biases garnered from artificial intelligence and deploy those biases in decision-making, resulting in biased judgments and discriminatory outcomes,³⁹ if the biases are left unchecked. Conversely, if the bias is towards fairness and equity, the benefit is increasingly empowering, unless blocked by harmful forces. These principles are true in education, employment and economic empowerment and in other spheres or aspects of human life. Thus, the fundamental solution is to ensure, from the inception, at the lowest rung of any digitalization and/or artificial intelligence, that there is a conscious effort to act for benefit and not harm, recognizing that once human agency is involved, there is going to be bias.

39. Ivana Bartoletti and Raphaële Xenidis, Study on the Impact of Artificial Intelligence Systems, Their Potential for Promoting Equality, Including Gender Equality, and the Risks They May Cause in relation to Non-Discrimination (Strasbourg, Council of Europe, 2023). ↩

In healthcare, housing, employment and education, data used as evidence to train artificial intelligence models have been prejudiced against people of African descent. First, by disproportionately reflecting specific demographics and second, by having certain assumptions or stereotypes about different groups embedded in the data. This means artificial intelligence models often reproduce and perpetuate the same inequities found in the sociopolitical and sociocultural contexts in which they are developed. Bad or incorrect data are not the only problem with the use of big data in socioeconomic decisions. The use of big data and algorithms in the context of healthcare, education, housing, employment and access to services, and in attaining cultural rights, has the potential to reproduce existing patterns of discrimination, inherit the prejudice of prior decision makers, or simply reflect the widespread biases that persist in society.

Through a racial discrimination lens, it is clear that the digital space is often misunderstood as disconnected from challenges that exist offline. However, a digital transformation agenda that does not take into account human rights will not bring meaningful change for people of African descent. Seemingly neutral technology has human rights implications. In developing regulations for new technologies, it is crucial for States to fulfil their human rights obligations. While ethics in digital transformation are important, it does not create obligations for States. It is therefore essential to maintain a focus on obligations of States under international human right treaties in order to fulfil the economic, social and cultural rights of people of African descent.

As the use of digitalization and artificial intelligence becomes more widespread in society, governments, educators, employers and citizens at large will need to pay more attention to the nature and quality of education, training and lifelong learning.

Digitalization and artificial intelligence will not affect all societies, all sectors of society or all strata of society equally. The known variables of level of education and level of poverty, demographic factors such as race and gender, and the nature of jobs all influence the impact.

The underlying issues of stable, clean energy, individual and societal levels of digital literacy, skills and competencies, the readiness of national and international leadership across all sectors, and the adequacy of legislation and policies with regard to harnessing digitalization and artificial intelligence for the public good must be addressed nationally and internationally.

International and national regulatory frameworks, namely conventions, legislation, policies and administrative procedures, still do not adequately address the current state of digitalization and artificial intelligence. Moreover, without considerable investment focused on closing the gaps, the necessary regulation of digitalization and artificial intelligence, and the monitoring of their impact, will continue to fall behind, to the detriment of traditionally disadvantaged groups.

The public discourse around, debate on and discussion of digitalization and artificial intelligence have not kept pace with the reach, scope and impact of digitalization and artificial intelligence. Similarly, education at all levels, from preschool to post-secondary education, does not adequately impart the knowledge required to ethically and equitably address the role that digitalization and artificial intelligence are already playing and will continue to play in all aspects of daily life. Thus, even among government personnel, knowledge of digitalization and artificial intelligence falls well below what is needed for governments to discharge their role in governance. The limited levels of digital literacy, skills and competencies across many societies makes many aspects of daily life vulnerable to the unplanned and unmanageable penetration of digitalization and artificial intelligence.

As the scope, reach and impact of digitalization and artificial intelligence continue to increase, there is growing consensus that human beings must remain at the helm and in control of these technologies. Such control must be maintained at the highest meta levels, in the interest of the good of humankind, and in every sector, at every level, in every organization and in every job, in the interest of individual and community well-being, and prevent breaches of the fundamental human rights and freedoms of every person.

Particular attention must be paid to emerging and evolving issues and concerns, such as whether robots should have legal identities and protections, and to the legal and ethical challenges that digitalization and artificial intelligence will engender. Also, attention must be paid to how the potential benefits of digitalization and artificial intelligence can and will be used, ethically and equitably, to address some of the perennial challenges that have haunted the human race, to reduce the divide between those who have and those who do not, and to prevent, contain and mitigate emerging and evolving challenges in the areas of environmental justice and digital justice.

Ultimately, these challenges require greater and more focused attention on and investment in those who risk falling behind, in the areas of governance, education and justice, as human priorities. To mitigate these concerns, the technology produced must be able to work within complex social realities, which require an understanding of legal, ethical and social contexts. This can only be achieved by incorporating more diverse perspectives into the technology sector.

Research on artificial intelligence has shown that there is broad agreement on the core content of the ethical principles that should be applied to artificial intelligence systems. The first, the principle of transparency, can be interpreted widely to include the explainability and interpretability of an artificial intelligence system; in other words, the possibilities for an individual to understand how the system works and how it produces its results. The second, the principle of justice and fairness, includes non-discrimination, impartiality, consistency and respect for diversity and plurality. It further entails that the subject of an artificial intelligence system’s operation should have the option to challenge the results, with the possibility of remedy and redress. The third, the principle of responsibility, encompasses the requirement that a human being should be responsible for any decision affecting individual rights and freedoms, with defined accountability and legal liability for those decisions. This principle is thus closely related to that of justice and fairness. The fourth, the principle of safety and security, posits that artificial intelligence systems should be robust, should be secure against outside interference and should include safeguards to prevent the performance of unintended actions, in accordance with the principle of precaution. The fifth principle is privacy. While respect for human rights generally might be considered inherent in the principles of justice and fairness and of safety and security, the right to privacy is particularly important wherever an artificial intelligence system is processing personal or private data.

States must raise general awareness of digitalization and artificial intelligence, ensuring a basic level of digital literacy in at least 87 per cent of the population, in line with the global adult literacy rate.

States must invest in public education on digitalization, artificial intelligence, e-governance, e-learning, e-health, e-finance and diverse e-applications, making the Internet accessible and available to all.

States must provide affordable and equal access to the Internet, to technological devices, to stable, clean energy sources and to lifelong learning for all.

States must invest more equitably in education, particularly computer education and digital literacy and skills at all levels, from preschool to post-secondary education, integrating an understanding of digitalization and the use of artificial intelligence in all disciplines and all spheres of life, as research shows that the return on investment in education is relatively high, as a public and a private good.

States must invest heavily in training for teachers of all levels, from preschool to post-secondary education, in the academic, professional and vocational fields, transforming their capacity to use and apply technology themselves and to teach their students to organically use and apply technology in their learning and routine activities.

States must institutionalize legislative and policy frameworks, informed by ethical principles and a commitment to fairness, equality and equity, to regulate the use of digitalization and artificial intelligence, generative artificial intelligence and the next generation of technologies, putting in place effective sanctions for non-compliance and establishing initiatives to mitigate and compensate for any harm caused by non-compliant entities.

States must support a recognition of the impact of race, racial bias and racial discrimination in digitalization and the use, misuse and abuse of artificial intelligence by supporting research regarding people of African descent.

States must ensure that digitalization and the use of artificial intelligence continue to be regulated, moderated and facilitated and that they remain subject to human oversight.

States must facilitate and incentivize the integration of people of African descent into economic development sectors, including employment, banking, finance and insurance, that are digitally progressive, ensuring that the existing divide is reversed rather than widened, deepened and prolonged, and facilitating enhanced innovation, research and development for and by people of African descent.

States should support research on the urgent and ongoing need to establish legal standards that ensure artificial intelligence is used ethically and in a manner that protects, promotes and fulfils human rights, for example in the justice sector and related sectors.

The Working Group urges States to include people of African descent in decision-making processes, to advocate for cultural responsiveness through intercultural frameworks that promote racial and cultural dignity to ensure a more inclusive experience for all, and to tackle racial discrimination directly and as a matter of urgency.

The Working Group urges States to ensure that all actors, State and non-State, that focus on the arts, sports and cultural expressions integrate human rights principles into digital transformation activities.

States should establish regulatory guidelines that require that face recognition and other artificial intelligence and language learning models include testing for racial bias and the publishing of results to increase transparency and accountability in the law enforcement and economic sectors and all other sectors using such technologies.

States should embed ethical considerations, particularly those that take into account health, housing and food security data and measures, into regulatory policies and practices.

States should ensure adequate funding and support for oversight bodies to monitor and evaluate the implementation of legislation, regulations and standards that govern the use of digitalization, artificial intelligence and new and emerging technologies, and to impose sanctions in the event of violations.