Federated Grassroots Community Model for Catalyzing Artificial Intelligence for Common Good Dr. Osamuyimen Stewart, Dr. Amir Banifatemi, Mathilde Forslund (AI Commons); Dr. Olubayo Adekanmbi, Olalekan Akinsande, Hamzat Oluwaseun, Halimah Oladosu, Wuraola Fisayo Oyewusi (Data Science Nigeria); Jumanne Rajabu Mtambalike, Essa Mohemmadali(Idea Labs); Timothy Kotin, Winifred Kotin, Emmanuel Odei, John Bagiliko (Superfluid Labs); Tara Chklovski (Technovation) Abstract The methodologies used in implementing and developing Artificial Intelligence initiatives and solutions for the common good are mostly deficient in achieving the true essence of the phrase “AI for Common Good”. In this paper, we are proposing a methodology for developing Artificial Intelligence solutions that adopt a grassroots federated community model involving a rare collaboration between problem solvers, end-users, and other stakeholders in identifying problems and conceptualization and development of Artificial Intelligence solutions in such a way that they can be accessible, reproducible, contextualized and incrementally enhanced across different locations making the possibilities of AI truly available to anyone, anywhere (AI for Common Good). Introduction The possibilities with Artificial Intelligence (AI) are immense and transformational (Liu et al., 2018). Unfortunately, it is still majorly a black-box model, which is difficult to access, reproduce, contextualize or enhance especially in poor and developing countries in Africa preventing these regions from truly benefitting from the possibilities and economic dividends that AI provides (Morris, 2020). AI initiatives and solutions developed by various experts and organizations to address this problem are often categorized or described as “AI for Common Good” (Berendt, 2019). Berendt (2019) identified the pitfalls and challenges of AI for Common God in the areas of problem-identification and solutionism mindset of the problem solvers, the difficulties of integrating different stakeholders, the role of knowledge, and side effects and dynamics after an exploratory study of 99 contributions to conferences on related fields. In this paper, we propose a new methodology for identifying and solving real and prevalent problems with AI and developing solutions to be made available to anyone, anywhere. Methodology The methodology adopts a federated grassroots community model that involves all stakeholders in the collaborative identification of problems and conceptualization and development of AI solutions via rapid prototyping. The problems and solutions are thoroughly documented, including the processes and learning to facilitate accessibility, reproducibility, and cross-border adaptation and enhancement of the solutions or datasets for incremental value creation through expanded collaboration networks. The proposed federated architecture is a method of solving complex problems by allowing interoperability and information sharing between autonomous or decentralized entities (Wikipedia, 2019). It has been implemented with outstanding results in crowdsourcing, crowd solving and gamified crowdsourcing models for problem-solving and driving innovation in business and other domains (Savage, 2020; Brabham, 2008; Geiger et al., 2011; Morschheuser et al., 2017) allowing for cross-fertilization of ideas and solutions from contributors from different locations and sectors (Gimpel et al., 2020) The proposed new AI for Common Good solution development methodology is currently being implemented by AI Commons in partnership with Data Science Nigeria, Idea Labs, Superfluid Labs, and Sahara ventures across 3 countries: Nigeria, Tanzania, and Ghana. We reveal some impressive outcomes of the methodology even in the difficult National lockdown caused by the COVID-19 pandemic demonstrating the effectiveness of the methodology even without any physical contact or collaboration. The project aims to demonstrate how the global community of AI experts can learn and co-create mutually beneficial solutions with the opportunity for incremental enhancements.