When slavery was abolished in the 19^th century, Trevelyan’s family “owned” over 1,000 people enslaved in 10 plantations across the small island. She believes she may have met some of their descendants when she visited Grenada for her documentary.

In April 2023, Trevelyan and seven others started a small group, Heirs of Slavery.

“We thought it was important to acknowledge that our ancestors had profited from slavery and to encourage others with this background to do the same,” she said. “The Caribbean has a 10-point plan, and we wanted to encourage Britain's government to engage in discussion with the Caribbean on the basis of this 10-point plan.”

More than 150 people descended from enslavers in the British Caribbean have contacted the group, Trevelyan added, including descendants of William Gladstone who was Prime Minister of the United Kingdom from 1868-1894.

Trevelyan pointed out that the issue of reparations is tackled differently in the Caribbean compared with the United States, where she lived and worked for more than 20 years.

“There's a compelling case for the idea of investment in health, education, infrastructure and climate resiliency, which is the Caribbean’s demand of the British debt owed,” she said. “Polling commissioned by the Repair Campaign shows growth in support for the idea of an apology and of reparations since the polling began — I think it's got to 40 per cent. Britain has now signed up for a discussion about reparatory justice for slavery with the Commonwealth. To me, that is the political space opening now is in Britain.”

At the session of the Permanent Forum, Trevelyan co-hosted a side-event with the Permanent Mission of Jamaica and descendants of enslavers in the Caribbean and South America, to discuss how the historical legacies of enslavement can be meaningfully addressed in the present. 

AI and digital justice

The Permanent Forum also hosted a discussion on digital justice for people of African descent. Artificial Intelligence (AI) is increasingly shaping critical aspects of modern life, including surveillance, employment, education, law enforcement, healthcare, data collection and analysis. Despite its potential to promote human rights, AI often inherits biases as it may be designed, used and regulated on the basis of structures that reflect power imbalances and that are discriminatory, and as the data it processes may lead to discriminatory outcomes.

For people of African descent, this presents a significant challenge, as they are frequently underrepresented or misrepresented in the datasets that inform AI systems. This results in the perpetuation of stereotypes and the deepening of racial disparities.

“The majority of research on race and technology comes from the United States however, the current administration has divested federal funding to universities. This means that, as technologies advance, we won’t get new knowledge,” said Mutale Nkonde. “Facial recognition systems are often used at borders and research has shown that people with darker skin are less likely to be recognized by these systems. The implications in terms of migration and asylum are huge.”

Nkonde is a researcher, policy advisor and media commentator advocating for policies and practices that reduce the expression of algorithmic bias in the design, deployment and governance of AI systems used in public life.

Her work began when she became the lead advocate for the Algorithmic Accountability Act, which was first introduced to the United States House of Representative in 2019, while she was a fellow at Data and Society. She consolidated this work after founding AI for the People, a global NGO that provides advisory services to major decision makers interested in responsible approaches to AI development.

Apart from vision, Nkonde offered speech as another problematic area of AI, supported by a paper by computer scientists at Carnegie Mellon University, entitled Annotators with Attitude, that found that social media users who wrote in African American Vernacular English were likely to have their speech tagged as toxic by AI driven content moderation systems. As a result, the paper elaborated, Black creators were less likely to reach their audiences.

“In the American context, using phrases such as ‘it’s the bomb’ or ‘it’s fire’ would be tagged as a threat, when in African American Vernacular these are simply ways of exclaiming delight, something joyful,” she said. Such contextualization, Nkonde added, pushes Black social media users out of a global content creation economy estimated at USD 250 billion by Goldman Sachs.

AI also offers the potential for positive change, particularly in the collection and analysis of disaggregated data, which could be instrumental in monitoring and addressing racial disparities, she added.

Nkonde highlighted medical breakthroughs thanks to AI that benefit Africans and communities of African descent, including pregnant women living in remote rural areas in Kenya where healthcare was scarcely available. Jacaranda Health partnered with tech giant Google to use AI as an additional tool to interpret images sent from portable ultrasound scanners, without the need for a sonographer. Nkonde also mentioned the 2024 Nobel Prize in Chemistry that showed how AI could be used for protein sequencing. However, the winners, Demis Hassabis and John M. Jumper, acknowledged ethical aspects that must be considered.

As several UN mechanisms have highlighted, the six principles of the human rights-based approach to data - namely participation, data disaggregation, self-identification, transparency, privacy, and accountability – should be applied to ensure data is used ethically, effectively and equitably.

For example, the Committee on the Elimination of Racial Discrimination’s General Recommendation 36 emphasizes the importance of preventing racial discrimination in AI applications and highlights the need for transparency, accountability, and human rights due diligence to mitigate the adverse impacts of algorithmic bias.

Nkonde, whose organization had previously collaborated with the former UN Special Rapporteur on contemporary forms of racism, Tendayi Achiume, on her report on racial discrimination in the design and use of emerging digital technologies, said that people of African descent need to advocate for their rights by engaging with global mechanisms such as the Permanent Forum and the UN’s other anti-racism mechanisms. She hopes the Permanent Forum’s submission for the elaboration of a United Nations Declaration on the promotion, protection and full respect of the human rights of people of African descent will include freedom from technology-enabled racism.

She also pointed out the need to interface with tech companies directly and increase racial literacy in the product development process. AI for the People has been sitting on the Content Moderation Advisory Board of TikTok’s US headquarters for the past five years.

“In 2025, AI for the People became shareholder activists, using their proxy votes to encourage the world's largest AI developers to adopt responsible approaches to AI,” Nkonde said. “One of our earliest victories was voting to maintain DEI policies at Apple.”