Responsible AI in farming: a multi-criteria framework for sustainable technology design

Mallinger, Kevin; Baeza Yates, Ricardo

doi:http://dx.doi.org/10.3390/app14010437

Responsible AI in farming: a multi-criteria framework for sustainable technology design

License

Citation

Mallinger K, Baeza-Yates R. Responsible AI in farming: a multi-criteria framework for sustainable technology design. Appl Sci. 2024 Jan 1;14(1):437. DOI: 10.3390/app14010437

Abstract

The continuous fusion of artificial intelligence (AI) and autonomous farming machinery (e.g., drones and field robots) provides a significant shift in the daily work experience of farmers. Faced with new technological developments, many risks and opportunities arise that need to be carefully translated into technological requirements to enable a sustainable production environment. Analyzing the complex relationship between social, ecological, and technological dependencies is a crucial step to understanding the different perspectives and systemic effects of technological functionalities. By providing a comprehensive overview of the state of the art, this article qualitatively analyzes the potential impact of AI on the autonomy of farmers and the technological developments to mitigate the risks. Fair data management practices, transparent AI approaches, and designs for an intuitive user experience are presented as key mechanisms for supporting responsible model development. Based on the defined social, technological, and ecological challenges in AI development, the knowledge to provide a high-level framework for the responsible creation of AI technologies is further systematized. By focusing on the multifaceted relationships and their effects on the autonomy of farmers, this article exemplifies the complex design decisions that must be faced in creating trustworthy and responsible AI tools.