A research team at the University of Veterinary Medicine, Vienna (Vetmeduni) has examined the potential of artificial intelligence (AI) in improving animal welfare in agriculture, highlighting both opportunities and ethical challenges. Led by Borbala Foris from the Center for Animal Nutrition and Welfare, the study found that while AI-supported systems can automate welfare assessment, significant hurdles remain in technology, ethics, and data quality.
Generative AI models, such as multimodal systems, show promise for monitoring farm animals at the individual level. However, current welfare protocols were designed for human observers and require adaptation for AI applications. A case study assessing the cleanliness of dairy cows demonstrated moderate accuracy, with 71% correct evaluations but notable weaknesses in precision.
The researchers stress that AI must not only improve efficiency but also reflect the real needs of animals. Ethical considerations and the One Welfare approach — integrating animal welfare, human well-being, and environmental sustainability — are central to ensuring responsible adoption. AI, the authors argue, should act as a “co-pilot” supporting farmers rather than replacing human expertise.
The findings were published in Frontiers in Veterinary Science under the title “AI for One Welfare: the role of animal welfare scientists in developing valid and ethical AI-based welfare assessment tools.”
Photo: Thomas Suchanek / Vetmeduni
Source: Vetmeduni
About the Author
