* Acknowledging that the categories of South and North are not watertight, this paper argues for situating geopolitical and geo-economic power within the history of post-colonial development.
The governance deficit that marks the AI paradigm today has produced a crisis of democratic and distributive integrity. It cannot be mended without a holistic vision that puts people and the planet at the centre, something that is not without precedent. Evolving this AI governance roadmap requires that we eschew both techno-pessimism and techno-fundamentalism. Blanket bans and knee-jerk reactions to AI may not be the solution (Schwartz & Sheard, 2021; Paz, 2021) and disproportionate attention to a hypothetical technological singularity (Lacker, 2021) can distract from the crucial narrative of power in the AI governance debate. At the same time, a business-as-usual optimism will not be sustainable.
What we have seen with the steady erosion of the civic-public space in AI-mediated sociality and the unjust order of the AI-enabled economy is inimical to a peaceful and just future. As the UN Special Rapporteur on Freedom of Expression (2018) has highlighted, the universal human rights framework can provide a robust starting point for evolving an effective AI governance approach. The needed shift, however, cannot just be a nominal rearticulation of global constitutionalism for the AI age. A global AI constitutionalism10 must carry the aspirations of multiple communities, privileging a future society where manipulation, loss of autonomy, exploitation and injustice have no place. It must be based on a rebalanced multilateralism for a renewed intelligence paradigm that sets a high standard for state obligations towards freedoms.
Humanity faces an emergency in the abuse of AI by unaccountable power – the weaponisation of dual use AI technologies, attacks on the sovereignty of state parties, and an unchecked aggrandisement of corporate power, particularly within the Global South. States must urgently embrace their duty to protect the rights of their citizens and uphold their extraterritorial obligations towards the human rights of all peoples. Beyond this, however, what is at stake is more than a state commitment. It is a willingness to act now to set the right foundations, build consensus on a framework for upholding rights in the age of AI – with adjustments in place to fill the gaps in international human rights law– and develop capacity to implement remedies.
Committing to a global AI constitutionalism is not the same as calling for a universal formula to address bias, discrimination, and inequality in the design of AI systems. A meaningful interpretation of the moral values that human rights represent in different contexts still needs to be adopted. For instance, privacy concerns emerge in unique ways across the Global South stemming from complex and situated notions straddling autonomy over personal information, the realities of shared use of digital artefacts, and communal identity (Ahmed et al., 2017).