AI helps decode emotions in seven animal species

A groundbreaking study from the University of Copenhagen has demonstrated that AI can decode emotions in animals. By training a machine-learning model to analyse the vocal patterns of seven ungulate species, including cows, pigs, and wild boars, the research achieved an impressive accuracy rate of 89.49%. This study, the first of its kind to cross species, marks a significant step forward in understanding animal emotions.

The AI model identified key acoustic indicators of emotional states, such as duration, frequency, and amplitude of vocalisations, revealing that emotional expressions are evolutionarily conserved across species. This discovery could revolutionise animal welfare, enabling real-time monitoring of animals’ emotional well-being, particularly in livestock management, veterinary care, and conservation efforts.

The implications for animal welfare are profound. Early detection of stress or discomfort could lead to timely interventions, improving animals’ lives. Additionally, promoting positive emotions could enhance overall welfare. The researchers have made their emotional call database publicly available to support further studies and encourage more AI-driven research in animal welfare and conservation.

This study not only sheds new light on animal emotions but also offers insights into the evolutionary roots of human language, opening up exciting possibilities for future scientific exploration and better understanding of animal behaviour.

For more information on these topics, visit diplomacy.edu.