The use of an AI depends on how it was trained. Are it subject to ideological constraints? Do its developers allow it to continue learning, or do they control the learning process and only allow certain things? And how smart are the developers themselves? An AI trained for Californian conditions is unlikely to be able to solve problems in the southern Sahara because the developers know nothing about them (and often, out of a certain arrogance, ignore everything). I think such AI is like the climate: it has to be locally anchored if it is to be useful (climate is also a local factor, even if this is always claimed otherwise for ideological reasons).

AI can significantly support sustainable development by optimizing resource management, improving agricultural practices, and enhancing energy efficiency, all while promoting digital inclusion and reducing technological disparities. By leveraging AI-driven solutions, communities can gain access to tailored educational resources and tools that empower marginalized groups, ensuring they participate in the digital economy. Moreover, AI can facilitate the development of low-cost technologies and platforms that connect underserved populations to essential services, such as healthcare and education. Ensuring equitable access to AI technologies fosters innovation and drives sustainable practices, ultimately contributing to a more inclusive and resilient global economy.

AI can support sustainable development by optimizing resource use, enabling smarter infrastructure, and advancing solutions in health, education, and climate resilience. However, to ensure digital inclusion and reduce tech disparities, it’s essential to invest in equitable access to digital infrastructure, capacity-building in underserved regions, and the development of transparent, bias-aware AI systems. Inclusive governance frameworks are key to making sure AI benefits are shared globally (UNESCO, 2023; World Economic Forum, 2022).