Ada Lovelace Institute hosts Global South civil society reflections on UN AI Dialogue

The Ada Lovelace Institute has published a first set of reflections from civil society experts on the UN Global Dialogue on AI Governance, whose first session took place in Geneva during July. Experts were invited by the Heinrich Boll Stiftung drawn from several regions including Africa, Asia and Central and South America, and six of them wrote forward-looking pieces ahead of the second meeting in New York in May 2027. The institute is releasing them in two blog posts during the 2026 UN General Assembly, with each piece presented as one expert's own opinion.

This post carries two contributions. Raymond Amumpaire, who works on AI ethics and governance in Africa, draws on a survey of participants in Geneva AI Week. Respondents raised four concerns: the gap between taking part and holding influence, the lack of routes from dialogue to implementation, interest in sovereign and local economic value, and the need for work between major meetings. Their hopes for concrete commitments were at best partly met, and participation was seen as tokenistic, with speakers from the Global Majority allotted three minutes. He also names sustainability, data centre resource use, labour conditions and human rights as neglected areas.

Mathew Lubari of Community Creativity for Development describes three groups at the event: corporate optimists, cautious diplomats and grassroots communities facing harms with little say. He argues the physical footprint of AI, including minerals, energy, e-waste and labour, receives too little attention. Both look to 2027 for stronger civil society channels, inclusive work between meetings, baseline standards across the AI value chain, repair rights, and greater participation by women, Indigenous Peoples and underrepresented communities. The institute notes the views are the authors' own.