Initiative overview
Brazil faces the challenge of providing clear, consistent, and easily accessible information on foreign trade procedures, as required under the WTO Trade Facilitation Agreement. For many years, this role was fulfilled through “Comex Responde”, a human-operated service that responded to questions from businesses and citizens. However, users had to submit forms and wait an average of four days for a response. This process led to delays, repeated inquiries, and difficulties navigating a complex regulatory environment, particularly for small businesses and individual traders.
To address these challenges, the Ministry of Development, Industry, Trade and Services (MDIC) launched “Comex Responde 2.0” and, in 2025, introduced Lina, an AI-powered virtual assistant integrated into Brazil’s national trade single window (Siscomex). Lina uses natural language processing to understand users’ questions and provide immediate responses, available 24/7. It relies exclusively on official sources, such as legislation, manuals, FAQs, and technical guidelines. When a response is insufficient or the issue is complex, users can request expert support and are redirected to the human service channel, ensuring continuity and accuracy of assistance.
The initiative aims to address three main challenges: long response times, repetitive questions that consume public administration resources, and the complexity of Brazil’s trade regulatory environment. By providing instant and standardised answers, Lina reduces uncertainty, enhances transparency, and lowers compliance costs for companies, customs brokers, and citizens, helping them meet regulatory requirements more efficiently. It also enables public servants to focus on more complex cases rather than routine and repetitive inquiries, thereby strengthening the overall quality of public service delivery.
Future development will focus on strengthening accuracy and governance. This includes continuous curation of the knowledge base, supervised learning based on real interactions, improved fallback mechanisms, and structured collaboration among technical departments and regulatory agencies. The goal is to institutionalise a permanent model of digital knowledge management within the Ministry. As the architecture matures, it may be expanded to other policy areas, allowing the platform to serve as a broader foundation for AI-supported public information services. By combining artificial intelligence with clear governance and official sources, the initiative enhances regulatory predictability and reinforces Brazil’s commitment to transparency in foreign trade.
