We are pleased to announce that the IRCAI Scientific Program Committees and the IRCAI Scientific Journal Editorial Board have completed their review of the Global Top 100 project submissions. 10 solutions were deemed “outstanding projects” based on their centrality of AI, potential impact on relevant SDG(s), demonstration of potential in completed work (either proof of concept or completed research paper ), and ethical design. Starting today, we will dedicate an article to each of these 10 outstanding projects to introduce them to relevant stakeholders and the broader public, and to make their voices heard on the world stage.
The first project labeled as “outstanding” that we will present to you is ASMSpotter – a tool that aims to “help local authorities effectively and continuously monitor artisanal and small-scale gold mining (ASGM) in large geographic regions by automating the detection of ASGM sites and applying Machine Learning and Computer Vision algorithms to satellite imagery.” The software was proposed as a basis to regulate the gold mining sector and plan ASGM formalization activities. In this way, the developers aim to enable local people to “have a share in the prosperity generated by artisanal gold mining and to minimize negative impacts on environment, health, and working conditions” (SDG12 + SDG15).
The team behind ASMSpotter emphasizes that “ASM is the source of livelihood for more than 44 million people in 80 countries around the world, and thus has enormous potential to contribute to the achievement of the SDGs.” At the same time, they recognize that ASM can pose a threat to human rights and the natural environment under a lack of governance. As such, the project aims to continuously monitor ASM as a way to address these challenges. “ASMSpotter provides an efficient and effective AI solution by combining cutting edge machine learning with specialist expertise on ASM through the partnership between Dida and Levin Sources.”
To prove the AI solution, ASMSpotter has been piloted on artisanal gold mining in Suriname: a convolutional neural network was trained on a dataset of more than 15,000 labeled satellite images from Planet Scope, created by experts at RWTH Aachen University. The team explains that “by automating the analysis of satellite imagery, ASMSpotter lowers the costs of monitoring and enables producing countries to more effectively govern their mining sector, enabling the protection of large geographical areas such as the Amazon”.