Light, algorithms and liability: A legal-optical risk index for AI-driven crop sensing under EU law

Authors

  • Constantinos Challoumis Department of Accounting and Finance, Faculty of Economics and Management, Philips University, Nicosia, Strovolos, Cyprus and Department of Business Administration, School of Economics and Political Sciences, National and Kapodistrian University of Athens, Sofokleous Str., 10559 Athens, and Department of Informatics and Computer Engineering, University of West Attica, Athens, Greece. https://orcid.org/0000-0002-7436-0506
  • Konstantinos Mavrommatis Department of Informatics and Computer Engineering, University of West Attica, Athens, Greece. https://orcid.org/0000-0003-2572-9247
  • Alexios Constantinou Department of Law, American University of Cyprus (AUCY), Ammochostou Avenue 52, 6019 Larnaca, Cyprus. https://orcid.org/0009-0006-0510-1881

DOI:

https://doi.org/10.55284/gjss.v12i2.2018

Keywords:

Agricultural law, Artificial intelligence, EU AI Act, Multispectral imaging, Precision agriculture, Quantification of Everything (Q.E.).

Abstract

Optical sensors carried aboard satellites, drones, and farm robots feed data into AI algorithms, which advise or apply plant treatments using fiber-optic and wireless communication. This research investigates the means by which the legal risk associated with such systems can be quantified prior to their implementation in European agriculture. Combining a doctrinal analysis of six layers of EU law (the AI Act as amended by the Digital Omnibus on AI, the GDPR, the Data Act, the Machinery Regulation, the revised Product Liability Directive and the NIS 2 Directive, along with the EU drone legislation) with the Quantification of Everything (Q.E.) method, the study develops the Legal-Optical Risk Index (LORI), with six weighted components, one of which is calculated physically from the ground sampling distance (GSD) of the camera. When applied to four illustrative crop production use cases, LORI ranges from 0.128 for the satellite-based vegetation index (NDVI) advisory system to 0.893 for the autonomous laser weed removal robot. The ranking of the scenarios proved robust in all 10,000 Monte Carlo weight variations. Legal risk in optical-AI agriculture is mostly governed by two engineering design decisions, namely, image resolution and machine autonomy, not by the AI model itself. Farmers, agricultural cooperatives, and providers of agricultural technology may employ LORI during sensor selection for the purpose of determining the optimal height of the sensor above the crops, image resolution, and the level of human oversight while mitigating legal risk.

Downloads

Download data is not yet available.

How to Cite

Challoumis, C. ., Mavrommatis, K. ., & Constantinou, A. . (2026). Light, algorithms and liability: A legal-optical risk index for AI-driven crop sensing under EU law. Global Journal of Social Sciences Studies, 12(2), 65–77. https://doi.org/10.55284/gjss.v12i2.2018