Using AI to monitor illegal wildlife trade

Biodiversity loss is one of the most pressing, yet often unrecognised, societal issues. Losing biodiversity and the benefits it provides to humans, in fact, also affects human wellbeing – for example, health and security in the face of climate change.

Illegal wildlife trade is one of the major drivers of biodiversity loss. Fuelled by the demand for body parts, such as ivory and rhino horn, illegal wildlife trade is threatening the persistence of iconic species, such as rhinoceros and elephant. In addition, it is also threatening the persistence of thousands of other species, mostly unknown to the public.

In the Information Age, illegal wildlife trade is increasing on digital platforms, especially on social media platforms such as Facebook. With an estimated two-and-a-half billion users, easy access has turned social media into an important venue for illegal wildlife trade. Wildlife dealers active on social media release photos and information about wildlife products to attract and interact with potential customers, while also informing their existing network of contacts about available products.

Dealers, for example, were found to offer endangered apes, such as chimpanzees, for sale on Instagram, in violation of international law. Currently, the lack of tools for efficient monitoring of high volume social media data limits the capacity to identify more such cases.

Finnish research is at the forefront of developing and using AI methods to efficiently monitor illegal wildlife trade on social media. Currently, we are developing machine learning methods that can process verbal, visual and audio-visual content in order to filter out information irrelevant to illegal wildlife trade from big data derived from social media.

We are also developing methods to classify the sentiment of social media users towards illegal wildlife trade. As these methods require labelled data to perform well, we are collaborating with experts from law enforcement agencies who have manually classified content pertaining to illegal wildlife trade, such as ivory, rhino horn and pangolin scales, on social media. Once the original information derived from social media is filtered, we will analyse the resulting data to better understand trends and patterns of this trade on social media.

Given the complexity of some of the issues being investigated, we have started collaborating with social media companies and other scientists working on artificial intelligence under the coordination of a global coalition to stop online illegal wildlife trade. The aim is to become one of the leading research groups globally investigating illegal wildlife trade online and to help stop the illegal wildlife trade on social media.

We are well on the way to achieving this goal.