ARtificial IntelligencE in fighting illicit drugs production and traffickiNg

Project Information

ARIEN

Grant agreement ID: 101121329

DOI

10.3030/101121329

EC signature date 1 August 2023

Start date 1 November 2023

End date 31 October 2026

Funded under

• Civil Security for Society

Total cost

€ 4 948 147,50

EU contribution

€ 4 268 160,00

Coordinated by

ENGINEERING - INGEGNERIA INFORMATICA SPA
Italy

Periodic Reporting for period 1 - ARIEN (ARtificial IntelligencE in fighting illicit drugs production and traffickiNg)

Reporting period: 2023-11-01 to 2025-04-30

Summary of the context and overall objectives of the project

The European Union’s commitment to combat illicit drugs trafficking has been reaffirmed through the EU Drugs Strategy and Action Plan 2021-2025. Despite the efforts and achievements of previous EU and international initiatives, trafficking of illegal drugs traffic is an increasing and dynamic phenomenon in terms of spreading and complexity, involving both organized criminal groups (OCGs) and lone criminals, who actively engaging together in a plethora of correlated crimes.
The requirement of specific knowledge to ensure that assessments which utilise tailored and tools to explore the emerging scenarios, threats and risks related to the different phases of drug lifecycle – also according to the specific category of substances - is becoming increasingly pervasive.
Criminal actors rapidly take advantage of new emerging technologies such as encrypted communications software and the opportunity to reach a large audience of consumers in complete anonymity offered by the Dark Web is being a common criminal modus operandi. The fast-changing supply chain and the evolving prototype of criminal actors can count on several direct and indirect crime enablers. Given the large availability of information and tutorials provided in the web and the accessibility of several kinds of technologies in the legal market, individuals who choose to engage in illegal activities are able to set up illegal laboratories for the production of illicit drugs in ordinary indoor civil facilities across the territory. Such unauthorised laboratories can count on drugs precursors and dual-sense chemicals readily available in the legal domestic market. The anonymity of the dark web infuses a sense of protection to user that are instead more exposed to cyber threats.
Financial flows and the large employment of cryptocurrencies get extremely difficult in the identification of suspected money transactions who the involved actors are from a distribution standpoint. Ultimately, this leads to criminals taking advantage of a dense network of mail and home delivery services to reach consumers.
In this complex panorama, LEAs and BGAs demand for novel technological support and improved regulatory actions to carry out drugs intelligence and investigative analysis in a co-ordinate and co-operative international setting capitalising on effective Artificial Intelligence (AI) techniques.

Work performed from the beginning of the project to the end of the period covered by the report and main results achieved so far

The ARIEN’s innovation action is organised into eight work packages (WPs). WP1 coordinates all the project activities and monitors its evolution.
An analysis of the social, ethical, legal and privacy aspects is conducted in WP2 to ensure the development of a framework compliant with EU laws and regulations in the field of drugs trafficking and completely respectful of human rights. To this end, guidelines, and checklists for applying legal and ethical The study conducted in the context of WP3 on criminal landscape and modus operandi of the entire drug supply chain, supported the analysis of the technological gaps of the security practitioners and their needs in fighting drug trafficking. This work package performed the elicitation of the user requirements and the definition of the use cases. Technical requirements have been collected and formalised in close collaboration with the representatives of LEAs, BGAs and Postal Service involved in the project. Such collaboration was also essential to specify the architectural design of the ARIEN Framework.
The development of the ARIEN framework, carried out in WP4, focused to enhance tools for discovering and monitoring both physical places (including parcels at border control) and online sources, including Surface and Dark web and social media, to discover and trace activities related to illegal drug cultivation, illegal production, and trafficking. Specific dark markets and web sites have been targeted by the ARIEN’s investigators to support the use cases selected for the project.
WP5 delivered the first version of the AI-based tools for detecting drugs clues and evidence within multimodal data (like texts, audio, image and video files) and to analyse crypto-currency transactions for discovering hidden networks of crypto actors. Furthermore, WP5 set the state for the creation of SELP datasets, designed to support the training of AI and ML algorithms.principles to the project activities and tools have been identified.
The initial version of the graph-based ARIEN Knowledge Base was implemented in the context of WP6 according to the data model defined in WP3. Dedicated applications for managing the chain of custody and the investigative reconstruction have been delivered for revealing insights and emerging trends, and for fostering cooperation among cross-border and cross-agency law enforcement.
Pilot scenarios have been defined and three demonstration rounds expected in the project have been planned (WP7). Dissemination and communication activities started at the beginning of the project with the participation to workshops and public events. The exploitation plan of the tools was initiated with the goal to innovate the technology currently used by LEAs and BGAs to fight the cross-country drugs trafficking phenomenon (WP8).

Progress beyond the state of the art and expected potential impact (including the socio-economic impact and the wider societal implications of the project so far)

ARIEN advances beyond the current state-of-the-art by leveraging AI for combating drug production and trafficking through several innovative tools, algorithms and methodologies. The project introduces advanced remote sensing solutions that accurately identify illegal cultivation sites using drone and satellite imagery, surpassing traditional surveillance approaches. Additionally, ARIEN employs novel shape-based object recognition in X-ray parcel scanning systems, significantly enhancing precision in detecting concealed narcotics and precursors in postal and customs environments. For secure and effective monitoring of illicit drug markets, ARIEN provides a sophisticated anonymous framework to extract targeted information from both Surface Web and Darknet. Furthermore, the project implements explainable AI methodologies, including multilingual named-entity recognition, visual semantic analysis, and robust trend analysis tools, elevating actionable intelligence extraction capabilities for law enforcement. Continued demonstrations in realistic operational settings and supportive regulatory frameworks are essential to validate the solutions and facilitate their broader adoption.

Arien-logo