Search for: All records

Direct exploitation, which includes the trade of wild animals for their parts, is a major driver of extinction. Digital communication tools, particularly the internet, have facilitated the trade in endangered species. Here, we automatically collected data to analyze online sales of threatened animals across 148 English-text online mar- ketplaces. We created a tool that searched for online sales of 13,267 animal species at risk of global extinction, as classified by the International Union for Conservation of Nature (IUCN), as well as 706 animal species on Ap- pendix I of the Convention for International Trade in Endangered Species (CITES), for which international commercial trade is prohibited. Examining a period of 15 weeks in 2018, we identified 10,699 unique listings selling body parts or eggs of threatened species, of which 4131 contained a full species name (common or sci- entific). These 4131 results were then filtered by keywords and, finally, manually vetted, which yielded 546 sale listings for 83 species. Of these 546 listings, 61 % advertised shark trophies (mainly jaws), 73 % of which were taken from species listed as endangered or critically endangered. Just four websites hosted >95 % of listings. We identified 18 species for sale that are included on CITES Appendix I. We also identified 13 species for which the IUCN had not identified intentional use as a threat. This work expands current understanding about the dealing of endangered and potentially illegal species online, specifies taxa threatened by online trade, and highlights emerging opportunities and persistent challenges to preventing the trafficking of threatened species. 

Wildlife trafficking remains a critical global issue, significantly impacting biodiversity, ecological stability, and public health. Despite efforts to combat this illicit trade, the rise of e-commerce platforms has made it easier to sell wildlife products, putting new pressure on wild populations of endangered and threatened species. The use of these platforms also opens a new opportunity: as criminals sell wildlife products online, they leave digital traces of their activity that can provide insights into trafficking activities as well as how they can be disrupted. The challenge lies in finding these traces. Online marketplaces publish ads for a plethora of products, and identifying ads for wildlife-related products is like finding a needle in a haystack. Learning classifiers can automate ad identification, but creating them requires costly, time-consuming data labeling that hinders support for diverse ads and research questions. This paper addresses a critical challenge in the data science pipeline for wildlife trafficking analytics: generating quality labeled data for classifiers that select relevant data. While large language models (LLMs) can directly label advertisements, doing so at scale is prohibitively expensive. We propose a cost-effective strategy that leverages LLMs to generate pseudo labels for a small sample of the data and uses these labels to create specialized classification models. Our novel method automatically gathers diverse and representative samples to be labeled while minimizing the labeling costs. Our experimental evaluation shows that our classifiers achieve up to 95% F1 score, outperforming LLMs at a lower cost. We present real use cases that demonstrate the effectiveness of our approach in enabling analyses of different aspects of wildlife trafficking.