Could AI save the world?

The environmental costs of data centres that power AI are coming under increasing public scrutiny given their impact on the environment, both in terms of energy consumption and use of water. However, if applied thoughtfully to the problem of environmental crime, there are unique opportunities for AI to contribute to protecting the natural world and supporting national security.

Environmental crime has emerged as one of the most significant transnational threats of the 21st century. INTERPOL and UN Environment Programme jointly assess environmental crime as among the fastest growing and most profitable crime types globally, ranking it as the fourth largest in the world by value [1]. This scale reflects not only the vast profits available to organised criminal networks but also the under prioritisation of environmental protection within national policing strategies.

Unlike conventional crime types, environmental crime often lacks an immediately visible victim. The harm unfolds slowly, diffusely, and across ecosystems rather than individuals. This temporal and spatial distance between cause and consequence means that law enforcement agencies, typically structured to respond to acute, person centred harms, are not designed to prioritise environmental offences, despite the potential to cause long term and irreversible harm.

The UK Government’s recent ‘Global biodiversity loss, ecosystem collapse and national security assessment’ outlined the problem in stark and unambiguous language [2]: "Critical ecosystems that support major global food production areas and impact global climate, water and weather cycles are the most important for national security....Ecosystem degradation is occurring across all regions. Every critical ecosystem is on a pathway to collapse." Both statements were given a 'high' analytical confidence rating.

The decline is further demonstrated in the WWF's Living Planet 2024 report [3]:

Environmental crime accelerates this decline by stripping natural resources, degrading habitats, poaching keystone species and empowering criminal networks that undermine governance.

Corruption sits at the centre of this threat landscape, as highlighted in The Financial Action Task Force (FATF), the supranational standard-setting body on money laundering, terrorist and proliferation financing [4], and United Nations [5] analysis, as a uniquely powerful enabler of environmental crime through the way in which it facilitates misallocation of funds, illegal licensing or disguises the illegal extraction of natural resources. The role of corruption in environmental crime is pivotal, as once an official grants a fishing permit, signs over a logging concession, or is paid to overlook a regulatory breach, the commodity and the entire downstream criminal supply chain can appear legitimate.

The global anti-money laundering (AML) framework places specific responsibility for the financial sector to apply enhanced due diligence (EDD) to individuals who have been entrusted with a prominent public function (politically exposed persons or PEPs) such as senior government officials, judicial or military officials and senior executives of state-owned corporations. Due to their position and influence, it is recognised that many PEPs are in positions that could be abused for the purpose of committing money laundering offences and related offences, including corruption and bribery.

The Regulations also require EDD to be applied to a PEP’s family or known associates offering a powerful but underutilised mechanism for detecting corruption linked to environmental crime. FATF supports environmental crime as a trigger for money laundering, yet current PEP controls are not systematically aligned with environmental risk indicators. Strengthening this alignment would allow financial institutions to detect suspicious patterns involving extractive industries, land use decisions, and resource allocation authorities that could indicate systemic money laundering activities.

Ben Brock has spent the majority of his career investigating transnational organised crime with a focus on financial intelligence, first for the National Crime Agency and then for TRAFFIC, a non-governmental organisation working to ensure that trade in wild species is legal and sustainable for the benefit of the planet and people. Building on his experience, combined with Principle One’s track record in implementing data analytics solutions within policing and national security organisations, we can set out a roadmap for applying AI to this problem.

Firstly, we can build a ground truth of environmental data, drawing on geospatial data, satellite imagery and habitat sensitivity maps to identify where extractive activity overlaps with ecologically critical zones. This consolidated spatial intelligence can become the foundation for the application of AI to identify organisations that operate in places where environmental harm would have the greatest national security consequences.

Next, we can begin to map human actors, drawing on public registries, beneficial ownership databases, extractive industry disclosures and trade records can provide a rich but fragmented picture of who is operating where. This is where artificial intelligence models can be applied to mine annual reports, environmental impact assessments, procurement documents and licensing decisions to extract the names of companies, subsidiaries, contractors and intermediaries involved in high impact sectors such as mining, logging, industrial agriculture and largescale infrastructure.

The capability of AI to consolidate multiple complex sources of data allows the construction of a network to highlight firms whose operations, supply chains or concession rights place them in sensitive ecosystems and enable more effective monitoring of their operations and environmental impact.

The political dimension is equally important, given the role that corruption plays in enabling of environmental crime. AI tools can be used to identify the governmental actors with authority over land allocation, resource licensing and regulatory enforcement, using tools such as graph-based machine learning models are to construct networks linking companies, beneficial owners, politically exposed persons, licensing bodies and geographic areas of operation. Patterns of proximity, such as repeated interactions between a high-risk company and a decisionmaker with control over concessions, can reveal corruption risks that would otherwise remain hidden.

By integrating data across these layers of ecosystem vulnerability, corporate behaviour and political exposure, AI could generate a continuously updated intelligence product. Sharing this with the global financial risk community would strengthen PEP monitoring, sharpen suspicious activity reporting, and ultimately help disrupt the corruption enabled environmental crime that threatens the UK’s ecological and national security foundations.

While much of this is easy to say, it does require a consensus to take action across a range of different sectors locally and on a global scale. Harnessing this knowledge through AI is just the first step and the reality is that despite the clear appreciation of the threat, the complex interplay between different government and problematic private sector actors is not easy to map out and understand. AI’s unique ability to sort and structure data from so many sources to identify the complex web of cause and effect offers an opportunity to demonstrate the role that investment in tackling environmental crime can play on both the financial and national security of the UK.  

[1] Nellemann, Christian. (2016). The rise in environmental crimes. A UNEP-INTERPOL rapid response assessment.

[2] UK Government. 2026. Global Biodiversity Loss, Ecosystem Collapse and National Security Assessment. London: Cabinet Office.

[3] Fig C, p21, WWF (2024) Living Planet Report 2024 – A System in Peril. WWF, Gland, Switzerland.

[4] FATF (2020), Money Laundering and the Illegal Wildlife Trade, FATF, Paris, France, www.fatf-gafi.org/publications/methodandtrends/documents/money-laundering-illegal-wildlife-trade.html

[5] United Nations Office on Drugs and Crime. (2023). Scaling back corruption: A guide on addressing corruption for wildlife management authorities. United Nations. https://www.unodc.org/documents/corruption/Publications/2023/23-12230E-Scaling_back_corruption_ebook_2023.pdf