1 INTRODUCTION

The rapid expansion of cross-border e-commerce, along with innovative business models such as online marketplaces that aggregate diverse sellers and products, coupled with the emergence of modern payment solutions and the global presence of individuals and businesses involved in both exporting and importing, presents a substantial challenge to the customs systems of many countries. The customs challenge is to seek measures that promote legitimate, secure, and compliant trade in this ever-growing and unstoppable international e-commerce landscape. With this focus, experiences from other countries, many of which have already been observed by Brazilian customs, such as noninvasive inspection, can greatly contribute.

Customs responses, tasked with both border protection and the expeditious movement of goods, need to be innovative and effective. Among the vast flow of global trade, a tool stands out as one of the most essential, if not the most important, for customs authorities to control the movement of goods: customs risk management.

2 AUTHORIZED ECONOMIC OPERATOR (AEO)

The AEO program is a trusted trader program that sets forth conditions for companies to voluntarily fulfil, enabling them to enjoy benefits in a trade-off arrangement. These benefits do not entail tax suspensions or exemptions as seen in special customs regimes, but rather stem from a stronger foundation of trust between customs and stakeholders.

As per Brazilian law, the AEO program was introduced through Regulatory Instruction No. 1.521, dated 5 December 2014. The program’s legal foundation is also established within the World Customs Organization (WCO) Framework of Standards to Secure and Facilitate Global Trade (SAFE) Framework, specifically under Article 3.32 of the General Annex.

The AEO program is designed to streamline trade processes and expedite the clearance of goods for trusted traders.

Approved OEA participants are entitled to several advantages, including reduced customs inspections, priority processing, and simplified documentation requirements. This approach is particularly beneficial for managing the surge in e-commerce shipments, as it allows customs authorities to concentrate their resources and stringent checks on higher-risk consignments and entities.

By focusing on the broader system, customs agencies can better target their resources, concentrating inspections and interventions on high-risk areas. This paradigm shift aligns with the complexities of today’s trade landscape and contributes to a more holistic, risk-based, and streamlined customs regime.

In this scenario, the advent of artificial intelligence (AI) has revolutionized various industries, including customs operations. In the context of customs, AI-driven automated risk analysis holds immense potential for enhancing efficiency and accuracy. By shifting from a transaction-based to a system-based approach, Brazil has modernized its customs treatment and controls. This transition involves a comprehensive assessment of the entire supply chain of a trusted trader, encompassing their security practices, financial solvency, and compliance history. This holistic evaluation fosters a higher level of trust between businesses and authorities, ultimately facilitating smoother trade flows.

The utilization of cutting-edge technology allows for the real-time evaluation of vast amounts of data, enabling quicker identification of high-risk shipments. This technology can swiftly assess transaction patterns, identify anomalies, and select consignments for closer inspection, all while expediting the clearance process for low-risk shipments.

3. ARTIFICIAL INTELLIGENCE (AI)

Brazilian efforts to implement computer-based procedures and controls began in 1993 with the launch of Brazil’s Integrated International Trade System (Siscomex). This system enables economic operators to record import and export transactions, contributing to a substantial database. Prior to the establishment of Siscomex, the absence of electronic data posed a challenge. Presently, the challenge lies in handling the vast and distributed data sets that are collected.

To aid its officials in efficiently extracting insights from this data and making informed decisions, Brazil’s Federal Revenue Service (RFB) has recently developed a trio of complementary intelligence tools: a machine learning-based tool, a desktop information gathering software, and a real-time risk monitoring system.

The AI tool is called System through Machine Learning (SISAM), a Portuguese acronym for ‘Customs Selection System through Machine Learning’ in English. This computerized tool has been in use since August 2014 to evaluate the risk posed by importations. It runs 24/7 in one of the Brazilian government’s data centres.

The AI tool ‘learns’ from the history of import declarations, both through supervised and unsupervised learning. Following the introduction of SISAM, customs inspectors are no longer obligated to examine the entire Declaration of Import (DI). Their responsibility is limited to reviewing the section of the DI that prompted its selection for the yellow, red, or gray channel. Additionally, inspectors have the discretion to choose whether or not to inspect other aspects of the DI.

SISAM’s supervised and unsupervised learning capabilities are not actually separated. They both emerge from the same probabilistic models and share the same knowledge base, which contains data associated with 8.5 billion different patterns.

SISAM calculates the return expectation of every possible inspection during the Customs clearance process. These expectations are later used to ‘feed’ decision and game theories, which support SISAM’s selection mechanism. SISAM has the ability to explain, in natural language, the reasons behind the selection of a cargo consignment and provide details on how it calculated the risk probabilities.

SISAM’s knowledge base can be updated incrementally, allowing it to learn from new import declarations every day without being retrained. The learning process can also be distributed across several machines.

4. ANOTHER IMPORTANT ADDITIONAL MONITORING SYSTEM IS PATROA

The Real Time Customs Operations Monitoring System (PATROA) was launched in December 2017, completing the current Brazilian IT ecosystem dedicated to Customs risk management. Just like ANIITA, it accepts human-created rules, but instead of acting only under user demand, PATROA runs server-side and applies the rules to transactions as soon as they are registered, thereby identifying risk profiles in real time.

PATROA can invoke ANIITA and SISAM to get detailed analyses of a transaction, then decide if an officer should be notified immediately or if further data should be collected for a later report. Brazilian customs authorities also frequently employ data mining methods on offline data, and are currently investing in a substantial data lake solution to centralize all data associated with both Customs and internal revenue services.

5. MUTUAL RECOGNITION AGREEMENTS (MRA’S)

Mutual Recognition Agreements (MRAs) have been signed among countries with the AEO Program, which mutually recognize each other’s AEO companies, granting them preferential treatment. In a significant development, Brazil and ten other countries signed the largest Regional MRA within the scope of AEO programs.

A key MRA was signed between Brazil and the US Customs and Border Protection after seven years of negotiations and validations. This agreement aligns Brazil's AEO with the US Customs-Trade Partnership Against Terrorism (C-TPAT). As a result, Brazil has drafted a new Regulatory Instruction (IN) that is under public consultation, with a deadline for implementation set for 1 July 2024.

6. CONCLUSION

The Brazilian customs landscape has undergone a transformative shift with the implementation of the SISAM system, providing customs inspectors with a more targeted approach to verification. This evolution is part of a broader trend towards leveraging technology to enhance risk management and operational efficiency. By incorporating data mining techniques and investing in a comprehensive data lake solution, Brazilian customs authorities are poised to create a centralized repository of information from both Customs and internal revenue services.

This forward-looking strategy not only streamlines processes but also positions Brazil as a proactive participant in the future of global Customs operations.