At the Captive Animal Enrichment Project (CAEP), one of our main practices is facilitating self-selection for animals. This novel approach to animal welfare is borne out of years of research and experience. But self-medication is only a small part of what we know about the animal kingdom, we still have much to learn.
We are enthusiastic about the emerging subject of artificial intelligence (AI) and its potential to transform our understanding of and approach to caring for animals in South Africa and worldwide, when it is correctly created.
Scientists have been studying animal behaviour, communication, and general wellbeing using conventional methods for decades. Although these techniques have produced insightful results, they can be labour- and time-intensive, and their applicability is sometimes restricted by human interpretation, as well as important elements which can often be overlooked by a person evaluating data.
How AI is currently being used with animals
AI has the power to completely alter the way we understand animal behaviour and communication and can improve the kind of welfare we currently offer animals in captivity. It is already revolutionising the process of deciphering the secret languages of the animal kingdom by utilising sophisticated algorithms and machine learning.
A Smithsonian Magazine story on the use of AI in research serves as an example of the fascinating potential. AI was used by a team of scientists to perform analysis of sperm whale vocalisations. Interpreting these complex clicks and whistles has always been difficult for humans. However, by using AI algorithms, researchers were able to distinguish between various “call types” that sperm whales use depending on the situation.
According to this ground-breaking research, AI may be able to shed light on animal communication patterns, social structures, emotional states, and even intricate networks within populations.
This work has enormous ramifications for the use of AI in animal research in the future. Imagine having the ability to “speak” with elephants, or understand what they are communicating, interpret the complex social dynamics of a pride of lions, or sense the emotional nuances of a chimpanzee’s cry. AI can help us to develop communications, leading to a greater degree of empathy and comprehension, and ultimately a closer relationship between humans and animals.
AI assisted camera traps, although limited in their development and not yet resilient and flexible enough to be adopted universally, are assisting in the detection of elephant herds approaching farmers’ crops, serving as a warning of impending crop raids and allowing farmers to drive the elephant away prior to any loss of resource, or ‘damage’ done to the elephant/human relationship.
AI is being used in fish farming in Norway to detect disease and improve treatments for health issues, as well as optimise breeding and genetics. Similarly Sainsburys, a major UK retailer, employs AI to watch their dairy cows 24/7 so that they can detect diseases early, monitor positive improvements in the behaviour of their cows as well as suggest ways to keep cows more comfortable to mitigate disease.
Successful trials in Portugal have led to implementation in Dutch wind farms of AI powered radar systems that detect flocks of birds approaching wind turbines, and stop them before collisions.
AI is being successfully developed to automatically recognise pain in various species, a welcomed helping hand in this notoriously difficult field of study, and yet crucial information for us humans to be aware of and keep in mind as we navigate our animal interactions.
While AI offers a less invasive way to study animals, removing entirely the stress evoked in animals due to human presence, it’s essential to maintain a balance between technological advancements and direct human interaction. If we become overly reliant on AI, we run the very real risk of becoming more detached from the natural world and hindering our ability to fully appreciate and understand animal behaviour. Integrative approaches are almost always likely to offer the most comprehensive outcomes.
How this impacts our work at CAEP
There are a number of fascinating uses of AI in zoopharmacognosy that we can foresee. Real-time animal behaviour monitoring is possible with the use of AI-powered cameras and sensors. This would enable us to see when and how animals seek out particular plants to self-medicate.
We can learn a great deal about the therapeutic qualities of plants and how animals use them to treat health issues by examining this data. AI is also capable of analysing enormous volumes of available animal behaviour data, finding correlations and patterns that human researchers could miss. New avenues for zoopharmacognosy may be discovered as a result, and our knowledge of how animals interact with their surroundings to be healthy may deepen.
Where else can AI be used?
AI has a lot more applications in animal care than only zoopharmacology. Algorithms driven by AI can be used to examine how animals move, identify symptoms of stress or disease, and even tailor enrichment activities to the needs of specific animals.
The wellbeing and quality of life of animals kept in captivity can be greatly enhanced by this individualised approach to care. Of course, there are ethical issues with using AI in research involving animals. It’s imperative to make sure AI applications are non-intrusive and are employed only after strict welfare measures have been put in place.
Of course it is up to the humans that program AI to take precautions against AI misinterpreting animal behaviour, the appropriate and ethical deployment of emerging technologies is of utmost importance. We think it’s a tremendous instrument that could completely change how we think about animals, and we support any technology that can contribute to our understanding of animal needs, behaviour and health so that we are able to develop more comprehensive and inclusive systems of welfare.
We can work toward creating a future where animals kept in captivity cope better with their environments by utilising cutting-edge techniques like zoopharmacognosy in conjunction with AI. We are dedicated to actively participating in animal welfare in an ethical and responsible manner to improve the future for all animals in South Africa.
If you are interested in keeping up with conversations involving AI and animals, as well as relevant resources and research on the topic, consider subscribing to the monthly newsletter from AI for animals by visiting their website on www.aiforanimals.org
Partnering with CAEP
As a non-profit organisation, we aim to enhance the lives of captive animals by using novel enrichment methods and zoopharmacognosy. To carry on with our important work, we depend on people like you for help. Here’s how you can help us:
- Donate – We are able to support research projects and provide vital resources for South African captive animals thanks to your kind donations.
- Volunteer – Provide your time and expertise to help us achieve our goals, whether it’s physical labour, financial expertise or knowledge about certain animal species.
- Raise awareness – Tell people about the value of animal welfare and the fascinating possibilities of AI in animal research by sharing this article.
We appreciate any and all forms of participation from the public and businesses. By working together, we can create a future in which all animals are treated with greater kindness and respect.
References:
Feighelstein, M., Henze, L., Meller, S. et al. Explainable automated pain recognition in cats. Sci Rep 13, 8973 (2023). https://doi.org/10.1038/s41598-023-35846-6
Brickson L, Zhang L, Vollrath F, Douglas-Hamilton I, Titus AJ. Elephants and algorithms: a review of the current and future role of AI in elephant monitoring. J R Soc Interface. 2023 Nov;20(208):20230367. doi: 10.1098/rsif.2023.0367. Epub 2023 Nov 15. PMID: 37963556; PMCID: PMC10645515.
