Frequently Asked Questions

Arbipedia is a research platform dedicated exclusively to arbitration in Brazil. We gather, organize, and classify by topic the relevant legal materials on the subject, so that research that would take more than 40 hours across scattered sources can be done in a few minutes, with greater reliability and breadth.

For lawyers, arbitrators, arbitral institutions, legal opinion writers (pareceristas), judges, universities, researchers, and students. The collection serves both those looking for a specific answer and those conducting in-depth research on an arbitration topic.

The platform brings together three distinct and complementary collections, all classified under the same topic structure — currently nearly 300 arbitration-related topics:

• National court case law — decisions of the Brazilian courts on arbitration;
• Scholarship catalogue — a catalogue of Brazilian publications on arbitration;
• Arbitral awards — decisions rendered in arbitral proceedings.

Each collection has its own logic of gathering, organization, and access, detailed in the following sections.

The case-law database gathers decisions of the state Courts of Justice (Tribunais de Justiça), the Federal Regional Courts (TRFs), the Superior Court of Justice (STJ), and the Federal Supreme Court (STF).

The inclusion criterion is thematic: it is not enough for a decision to mention arbitration in passing — it must actually address a relevant arbitration issue.

We systematically monitor the decisions published by these courts and select those dealing with arbitration. Gathering is continuous, and the collection is updated weekly.

Yes. Court decisions are public acts, and the platform hosts the full text of each ruling: just click on the decision to open the complete text.

This is the key to understanding how our case-law research works, and how it differs from what you find in the courts' own search engines.

We use the word "headnote" (ementa) for a statement drafted by Arbipedia that summarizes a decided arbitration-related issue and gathers, under it, the similar rulings that addressed it. This headnote is not the ruling's official headnote (ementa): it is a point of thematic aggregation, written by our team, that groups the decisions that examined the same issue.

So, when searching, users do not receive a raw list of isolated rulings. They receive the identified legal issue and, beneath it, the set of decisions that decided it — which makes it possible to see at once how the topic is being treated, where there is convergence, and where there is divergence.

Because traditional keyword-based case-law research brings a lot of noise (decisions that merely mention the term) and misses relevant decisions that use different wording. Organization by headnotes solves both problems: the work of identifying, classifying, and grouping has already been done by people who know the subject. Each headnote connects to the subject index and to the corresponding article of the Brazilian Arbitration Act (Law No. 9,307/96), making it possible to go from the statutory provision directly to the case law interpreting it.

It is a catalogue of Brazilian publications on arbitration — books, chapters, journal articles, and other academic and professional works. We gather and classify the references, with the bibliographic data, the table of contents when available, and an indication of the related topic.

The goal is to answer the question that precedes any doctrinal research: "what has already been written about this in Brazil?"

No, and this is a deliberate choice. Third-party publications are works protected by copyright, and reproducing them in full would require authorization from the authors and publishers. Arbipedia strictly respects those rights.

What we offer, as to third-party scholarship, is the catalogued reference: the data that allow users to locate the work, assess its relevance, and obtain it through legitimate channels — the publisher, the journal, the library, or the corresponding repository.

Not necessarily. The catalogue tells you that the work exists and what it deals with, not the thesis the author defends. Attributing a position to an author based solely on the title of the work would be inaccurate and, at times, unfair to the author. When the platform has information about the content — a table of contents or a description — it is presented as such, and not as the author's opinion.

Yes. Arbipedia produces its own original publications, and subscribers have full access to these within the platform.

This is the essential distinction of the scholarship collection:

We follow Brazilian scholarly output on arbitration and catalogue the publications we identify, assessing and classifying them under the same topics that structure the case law. This means that research on a given issue returns, side by side, the corresponding decisions and bibliography.

Authors and publishers may suggest publications for cataloguing by email at This email address is being protected from spambots. You need JavaScript enabled to view it.. Including the reference does not involve reproducing the text and imposes no cost on the author.

It brings together decisions rendered in arbitral proceedings — material that has historically been hard to access, given the confidentiality that usually attaches to arbitration, and that is a valuable source for understanding how issues are actually being decided in arbitral tribunals.

• 4.2.1 Why am I seeing a full-text arbitral award on Arbipedia?

  The arbitral awards made available on Arbipedia were obtained from public court proceedings. When an arbitral award is submitted to the courts, it becomes part of the case record and, as a rule, is subject to the constitutional principle of open court proceedings (publicidade processual), under art. 5, LX, and art. 93, IX, of the Federal Constitution.

  The public nature of these awards therefore stems from the legal regime applicable to the underlying court proceedings, not from any act of Arbipedia. In making these decisions available, Arbipedia processes information that is already public for the purpose of organizing and providing knowledge about Brazilian arbitral case law, contributing to legal research, transparency, and legal certainty.

  This processing is grounded in art. 7, IX, of Law No. 13,709/2018 (the Brazilian General Data Protection Law — "LGPD"), subject to the parameters of § 3 of the same article, under which the processing of personal data whose access is public must consider the purpose that justified its disclosure, good faith, and the public interest. In concrete terms:

  • Purpose: Arbipedia uses the awards for legal research and to organize a knowledge base on Brazilian arbitration.
  • Good faith: Arbipedia takes measures to limit the exposure of personal data, including anonymizing the parties' identifying data and providing a channel for requests relating to personal data.
  • Public interest: the organized availability of arbitral decisions promotes legal research, transparency, and legal certainty in the development of Brazilian arbitration.
• 4.2.2 How does Arbipedia handle the data contained in arbitral awards?

  Arbipedia adopts different measures depending on the nature of the data and its relationship to the intended purpose, observing in particular the principles of purpose, adequacy, necessity, security, and prevention set out in art. 6 of the LGPD:

  • Parties' identifying data: names, identification documents, and qualifying information are anonymized before the award is made available, so as to avoid exposing information unnecessary to the purpose of researching and organizing arbitral case law. Because of the anonymization, such data are not considered personal data for the purposes of the LGPD, under its art. 12.
  • Content of the decision: the legal content of the award is preserved — the facts, arguments, legal reasoning, and the arbitral tribunal's decision — as these are elements necessary to understand and research the decision. The anonymization of the parties' data does not extend to this information. In this way, Arbipedia seeks to reconcile the protection of personal data with the preservation of the legal content necessary for research.
  • Identification of the arbitrator: the arbitrator's name is retained. The platform acknowledges that this is personal data, but its retention is justified by the functional nature of the information: under art. 18 of Law No. 9,307/1996, the arbitrator is a judge of fact and of law, and their name, associated with the award they rendered, identifies who exercised the decision-making function and the composition of the arbitral tribunal.

  This understanding is supported by the case law of the Federal Supreme Court. In the judgment of SS 3.902 AgR-segundo, the STF distinguished information relating to the individual as a public agent from information belonging to their private sphere, recognizing that functional data may be subject to disclosure when linked to the exercise of the function. In ADI 4.414, the Court held that the public nature of procedural acts is the rule and that blanket restrictions on access to the case file are incompatible with the Federal Constitution. This understanding is reflected in CNJ (National Council of Justice) Resolution No. 647/2025, which expressly provides for reconciling personal data protection with the public nature of proceedings, allowing access to and sharing of personal data for transparency purposes.

  In comparative law, the Court of Appeal of England and Wales, in Tickle & Anor v The BBC & Ors [2025] EWCA Civ 42, overturned an order that had directed the anonymization of judges sitting in court proceedings, recognizing that identifying those who exercise a judicial function is inherent in the principle of open justice and essential to public scrutiny of the justice system. The Court held that only concrete and specific evidence of serious risk could, in wholly exceptional circumstances, justify such a measure.

  When an arbitral award forms part of public court proceedings, the arbitrator's name — functional data associated with the authorship of the decision and placed in a context of open proceedings — is retained on Arbipedia for the purposes of legal research and transparency of the arbitration system.

• 4.2.3 How do I remove information contained in the arbitral awards?

  The parties' identifying data are anonymized before publication. If you identify any personal information that has not been properly anonymized, you may request its review through the channel indicated below.

  As to the other elements of the award, requests for removal or alteration will be analyzed individually, considering the nature of the information, its relationship to the purpose for which the decision is made available, the existence of a legal basis for the processing, and the data subject's rights under the LGPD.

  A mere removal request, therefore, does not result in the automatic takedown of the content.

  If there is a subsequent court decision ordering the confidentiality of the underlying proceedings or of the arbitral award itself, Arbipedia may reassess the availability of the document upon presentation of that decision.

  For clarifications, requests, or applications relating to the processing of personal data, contact us by email at This email address is being protected from spambots. You need JavaScript enabled to view it..

In addition to the index (filtering the various topics), the platform offers free keyword search, with operators that refine the results.

In the results, each item is tagged by collection — CASE LAW, SCHOLARSHIP, or ARBITRAL AWARDS — which lets you tell the nature of each source at a glance.

For a specific question, a keyword usually suffices. For a broader survey — a legal opinion, a memorandum, an article, preparing a defense — we recommend the subject index, precisely because the thematic organization reveals material that a text search would miss. To follow a specific author, the author index is the direct route.

Yes. For the platform's original content, we suggest the following format:

RANZOLIN, Ricardo (coord.); GALAN, Lorenzo; WERLE, Vitória (eds.). Arbipedia. [title of the material]. Arbipedia, Porto Alegre, [year]. Available at: [URL]. Accessed on: [date].

For third-party case law and scholarship located through the platform, cite the original source — the ruling or the publication — and not Arbipedia. We remind you that the platform's material may not be published, rewritten, redistributed, or transmitted without citing the source.

No. Arbipedia's content is reserved for subscribers: access to the collection and to the full text of the material requires a subscription.

The available plans, their prices, and the subscription terms are on the Subscriptions page. If you have any questions, write to This email address is being protected from spambots. You need JavaScript enabled to view it..

It is an AI-assisted search tool that answers natural-language queries based exclusively on Arbipedia's collection. Instead of browsing the indexes, users ask a question and receive a structured answer, with an indication of the sources that support it.

No. arbiped AI was built to answer only from the material actually found in the collection. When the database does not contain enough information about the query, the tool tells the user so — instead of filling the gap with general knowledge, which is the main source of error in AI tools applied to law.

The tool is expressly instructed not to produce case numbers, party names, courts, dates, headnotes, or scholarly references that do not appear in the retrieved material. Every statement must be traceable to a document in the collection, and the sources are indicated in the answer itself, with a link, for direct verification.

Even so, we recommend that users verify the indicated sources before using them in a court filing or academic work. The tool speeds up and broadens research; it does not replace the professional judgment of the person using it.

The case-law collection is updated weekly. The academic commentary and arbitral-award collections are expanded continuously, though without the same fixed schedule.

To keep up with what's new — new rulings, publications, and materials — we recommend subscribing to our Newsletter, through which we send the platform's main updates. It's worth subscribing so you don't miss what is added to the collection.

Meet the team, the coordination, and the editorial project on the About us page.

By email at This email address is being protected from spambots. You need JavaScript enabled to view it..