Banca de QUALIFICAÇÃO: JOÉLCIO DE CARVALHO TONERA

Uma banca de QUALIFICAÇÃO de MESTRADO foi cadastrada pelo programa.
STUDENT : JOÉLCIO DE CARVALHO TONERA
DATE: 30/05/2025
TIME: 10:00
LOCAL: https://meet.google.com/jwz-cwgm-kqq
TITLE:

APPLICATION OF ARTIFICIAL INTELLIGENCE IN TRADEMARK REGISTRATION: A METHODOLOGY TO SUPPORT PROFESSIONALS IN FILING WITH THE INPI

KEY WORDS:

AI Technologies in Intellectual Property. Trademark Registration. Prompt Engineering. Regulatory Management of Intellectual Property.

PAGES: 63
BIG AREA: Ciências Sociais Aplicadas
AREA: Administração
SUMMARY:

This master's research aims to systematize methodological guidelines for the application of Artificial Intelligence (AI) technologies in the regulatory management of Intellectual Property assets, with a specific focus on reducing formal and substantive errors in trademark applications filed with the Brazilian Patent and Trademark Office (INPI). The investigation is based on the premise that Large Language Models (LLMs), when combined with Retrieval-Augmented Generation (RAG) techniques and Prompt Engineering strategies — and properly parameterized using a structured normative corpus — may contribute to greater accuracy and efficiency in the filing process by providing legally sound and contextually grounded answers.

The research does not intend to develop new software, but rather to generate applied technical-scientific knowledge through a guiding technical report, which will describe the methodological steps, operational safeguards, and evaluation criteria relevant to the strategic use of these technologies in the field of trademark registration. The resulting product is intended to assist legal professionals, innovation managers, and IP practitioners in structuring and parametrizing regulatory databases, formulating optimized prompts, selecting appropriate AI architectures, and assessing the performance of interactions, in order to avoid misinterpretations or regulatory inconsistencies.

The execution of the project is organized into four main phases. The first, already completed, involved a technical-scientific review of applicable technologies, focusing on LLMs, RAG, and Natural Language Processing (NLP). The second phase, currently underway, comprises the curation, semantic structuring, and standardization of INPI’s normative corpus, including legislation, manuals, resolutions, technical guidelines, and administrative instructions. The third phase will involve experimentation with prompt engineering strategies and different retrieval and response methods, aimed at assessing the interpretive effectiveness of AI in simulated contexts. Finally, the fourth phase includes methodological validation through qualitative analyses and interviews with experts and potential users from the legal and innovation ecosystems.

At the end of the project, a single technical product will be delivered: a Technical Report on Intellectual Property and Technological Innovation, presenting the theoretical foundations, methodological procedures, evaluation results, and practical recommendations for the replicable use of the proposed methodology. This report is expected to guide legal professionals, innovation hubs, IP offices, and public institutions on the safe and effective use of AI technologies in supporting trademark application procedures before the INPI.

COMMITTEE MEMBERS:
Presidente - RODRIGO PEROZZO NOLL
Interna - GENIZIA ISLABÃO DE ISLABÃO - IME
Externa à Instituição - GABRIELE SARMENTO DA SILVA
Externo à Instituição - KATYUSCO DE FARIAS SANTOS