Using AI and Big Data to Reduce Global Illegal Trade in Plants

The survival and prosperity of humanity are closely related to the conservation and utilization of plant diversity. Of the approximately 390,000 species of higher plants known worldwide, about 50,000 are utilized by people, and the main economic crops cultivated in various countries have been fixed during the global introduction and domestication process. However, plant diversity is facing threats worldwide largely due to human impacts [1]. About 21% of plant species are threatened [2]. International conventions such as the Convention on International Trade in Endangered Species of Wild Fauna and Flora (CITES) and the Convention on Biological Diversity (CBD) are set up to protect, restore, and sustainably utilize diverse plants and biodiversity in general. Plants also figure prominently in the United Nations Sustainable Development Goals (SDGs). By the end of the fifteenth century, new shipping routes connected parts of the world that had previously been isolated. Ever since, human society has been on a trajectory toward integrated global development, which has included the introduction and promotion of crops such as potatoes and corn, and the increasing yield and diversity of agricultural products. This has brought about significant cultural, social, and economic changes, especially since World War II, when global trade and the global transportation network expanded dramatically. However, the volume of the illegal trade of wild plants has also increased since the 1990s′, including through sophisticated cross-border networking and concealment, which has created regulatory challenges. Global trade has a significant role in the overexploitation of wildlife, and the diversity of species traded across borders continues to increase. Although global connectivity has brought economic benefits, it has simultaneously created new challenges for biodiversity conservation. Over 180 countries have signed the CITES convention, which regulates and monitors over 30,000 plants threatened by international trade. The current release of the World WISE Database maintained by the United Nations Office on Drugs and Crime records nearly 180,000 seizures in the period 1999 to 2018 from 149 countries with almost 6000 species represented in the illegal trade [3]. No species accounted for more than 6% of total seizures, and no country accounted for more than 15% of seizures. Between 2016 and 2020, global exports of CITES Appendix plant species amounted to 9.3 billion US dollars [4]. These data highlight the severity and diversity of illegal trade in wildlife, and demonstrate that this is a global issue. United Nations Conventions including CITES and the Convention against Transnational Organized Crime, provide a legal basis for international cooperation in response to reducing the global illegal trade of wild plants. National governments have also recognized the serious impact of this illegal trade and have committed to combat this issue through international cooperation and domestic legislation [4]. The International Union for Conservation of Nature (IUCN) and Botanic Gardens Conservation International (BGCI) have also launched global information campaigns to highlight the increasing illegal trade in plants. It is now widely recognized that illegal trade in plants needs to be addressed with greater vigor because of its impacts on endangered species and ecosystems, its detrimental effects on livelihoods and traditional cultural knowledge, and the loss of revenue for governments. Despite considerable efforts from governments and civil society, the current systems for regulating and curtailing the illegal global trade in plants are struggling in the face of increased illegal cross-border trade. Current efforts are clearly insufficient. Traditional measures to combat illegal trade in plants mainly rely on passive methods such as customs inspections, field patrols, and market spot checks, which are inefficient and have blind spots in monitoring, making it difficult to deal with dispersed and concealed online transactions and transnational criminal chains [5]. By utilizing artificial intelligence and big data innovative technologies such as intelligent monitoring and prediction [6], precise tracking and evidence chain closure, global coverage and collaborative efficiency can be achieved, relying on data-driven decision-making, real-time analysis, and multi-dimensional collaboration, to achieve full cycle and intelligent prevention and control of illegal plant trade, significantly improving the accuracy of crackdown and governance effectiveness. The growth of the Internet has provided new opportunities for promoting global trade, which has also facilitated illegal trade in plants and plant products. Improved capabilities to use AI and big data technology offer a new opportunity to improve the efficiency of which illegal trade in plants can be addressed. Advances in AI applications for big data analysis have improved the efficiency of search and analysis technology [7]. Automating the collection, analysis, and application of protection and trade data for threatened plants, using AI and big data technology can significantly improve the detection and prevention of illegal trade, with the prospect of significantly improving efficiency that offers the promise of achieving greater efficiency with reduced resource expenditure [8]. One example is The Global Alliance to Combat Illegal Trade in Wild Fauna and Flora launched by the World Wildlife Fund, the International Wildlife Trade Research Organization, and the International Fund for the Conservation of Animals in 2018. Between 2018 and 2023, the 47 members of the Global Alliance deleted a total of 16.07 million posts or advertising messages suspected of connection to illegal wildlife trade. Through the Alliance's “Online Wildlife Crime Investigator Program,” information on over 23,000 items of illegal trade in wildlife have been discovered and reported. Real-time automated species level detection of trade document systems that assess shipment declarations and invoices to collect data on species being traded [9], FloraGuard [10], and China's Smart Goalkeeper Customs System (Figure 1) are other examples of the application of AI and big data approaches in reduced detection time and increased accuracy rate. Applying artificial intelligence and big data technology to reduce global illegal trade in plants by artificial intelligence and big data technology. These interventions show that AI and big data can improve the effectiveness of protection for threatened plants by analyzing the purchasing behavior of buyers in websites that sell plants, understanding their requirements and preferences, and analyzing their locations, which allows for more targeted sales blocking strategies. Real-time monitoring of patterns of transactions also allows illegal trade to be more effectively blocked and disrupted [9]. Such actions also allow the identification of suppliers, websites, and buyers engaged in criminal behavior, which can then be addressed by law enforcement. More generally, improved data on the scale and scope of illegal trade in plants, especially the identification of those sectors where such trade is concentrated, allows improved awareness raising efforts that can be leveraged to promote protective behaviors. Education and outreach, and behavior change are crucial to reduce illegal trade in wild plants by guiding consumers toward legal products. The use of AI and big data technology has an important role to play, especially when supported by DNA-based methods for rapid identification and determination of sources and legality. Accelerating the use of AI and big data technology to reduce illegal trade in plants and plant products will require international cooperation to establish an information sharing platform that can coordinate enforcement action at all levels to sever the links in the chain of illegal trade [10]. At the same time, improving regulations and policies, optimizing certification procedures, enhancing coordination within and among law enforcement organizations, and strengthening online monitoring and tracking of illegal wild plant transactions will all be helpful. Further, because the regulatory understanding of AI varies internationally, clear checks and balances also need to be established to manage security risks [11]. Privacy protection and the avoidance of bias and intellectual property issues in training data will also be required to ensure informed human oversight. Artificial intelligence can also simulate brain-like calculations through the simultaneous application of knowledge, data, algorithms, and computing power to connect perception and cognitive systems [12]. The potential applications for using AI and big data to reduce illegal trade in threated plants show significant promise but require specific new investments in four areas. (1) Build a cross level data and AI infrastructure network. Integrate plant import and export data from global, regional, and national levels of commercial trade, and data from zoos, botanical gardens, and institutions, establish standardized data centers and AI technology platforms, and provide basic data support and technical foundations for illegal plant trade monitoring. (2) Research and develop intelligent identification and collaborative monitoring systems. Integrate computer vision and natural language processing to develop cross platform monitoring tools that support dark web crawling, establish cross-border and cross institutional training data sharing libraries (including legal and illegal trade samples), realize automated monitoring of illegal transactions through e-commerce, social platforms, and other channels, real-time application of multi-agent data sharing and analysis results [13], and establish a database of species DNA barcode and image feature association. (3) Establish a collaborative law enforcement and intelligent decision-making system. Build a networked sales dynamic review system that integrates satellite remote sensing (such as identifying illegal planting areas), supply chain map analysis, and real-time threat scoring models, outputs hierarchical law enforcement suggestions, and develops intelligent decision-making tools for government law enforcement [14]. (4) Formulate ethical standards and technical guarantee mechanisms. Establish ethical standards for AI and big data in the review of illegal plant trade, establish a three-tier audit mechanism including algorithm bias detection, law enforcement impact assessment, and cross-border judicial compatibility review, and generate court reports that comply with the International Standards for Digital Evidence. In utilizing artificial intelligence and big data to reduce global illegal plant trade, there are also governance and ethical issues that need to be addressed. (1) The obstacles to cross-border compliance of data are due to differences in data sovereignty policies among countries, leading to issues with the legitimate capture of e-commerce/social media data. (2) The lack of unified standards leads to inefficient cross-border law enforcement cooperation, and legal differences between different countries and regions hinder cross-border law enforcement cooperation. (3) The credibility issue of judicial evidence requires the development of interpretable AI layers to ensure transparent and traceable algorithm decisions and meet the requirements of the court evidence chain. (4) The legal issue of protecting personal privacy when collecting and using data. (5) The issue of fairness risk in algorithms, where training data may contain regional or species biases, requires the establishment of a third-party fairness audit mechanism. (6) Potential risks of technology abuse, such as criminal groups potentially countering AI monitoring. (7) How to make the public understand and trust the regulatory process involving artificial intelligence. The expansion of global trade since the sixteenth century has changed patterns of global trade with concomitant impacts on the global population, the global economy, and socio-cultural organization. Key interventions to reduce demand are improved education and strengthened breeding and production technologies that provide legal commercial sources. By themselves, however, these interventions are unlikely to be sufficient. Integrating and combining AI and big data technology to prevent and control illegal plant trade offers new tools and approaches that can change the rules of the game for the protection of threatened plants. This study provides the first comprehensive framework for integrating AI technologies across the entire plant trade monitoring pipeline, from detection to enforcement. In the practical process, there are still issues such as data privacy concerns in AI monitoring, potential algorithmic biases in plant identification, and challenges of international data sharing. Hai Ren: conceptualization, funding acquisition, writing – review and editing, methodology, writing – original draft, visualization. Yifu Wang: writing – review and editing, conceptualization, visualization. Stephen Blackmore: conceptualization, writing – review and editing. This work was supported by the Guangdong Province Basic and Applied Basic Research Project (Grant No. 2023B0303050001). We thank Sir Peter Crane FRS for his valuable comments. Thanks also to the anonymous reviewers for their constructive comments.