L
LawX Brazil
LawX is an AI case workspace for Brazilian lawyers, sold by ADVBOX, the Florianópolis legal software company that bought the Brazilian LawX brand and assets in 2026. It reads a case's documents, audio and WhatsApp conversations, builds a fact memorandum and a sourced timeline, marks each fact as confirmed or alleged, and flags contradictions and gaps in the evidence. It then ranks possible legal actions with a confidence level and links each one to STJ, STF and state court precedents.
Agents cover practice areas such as social security, labor, family, tax, banking and administrative law. The social security agent works from contribution records and expert reports and, in one example, weighed 36 possible actions and suggested three. Drafting runs inside ADVBOX, where a lawyer picks a strategy, edits the generated filing and exports it, while the standalone LawX site still lists filing generation as coming soon.
The free Petition Reviewer rates a filing from 0 to 10 on six criteria, quotes the passage behind each comment and checks cited precedents against LawX's own collection. Paid agents cost R$200 each a month, with access rolling out in stages.
Alternatives to LawX Brazil
Other products that do the same kind of work as LawX Brazil, and what each one does.
- AArgus LegalArgus Legal is an AI platform for Argentine lawyers, run from Córdoba with teams in Valencia, Santa Monica and Las Condes and founded by Franco Giavedoni, a lawyer, and Luciano Giavedoni…
- AAranzadi AllegraAranzadi Allegra is the generative AI legal assistant of the Spanish legal publisher Aranzadi LA LEY, working from the publisher's own collection of close to 10 million documents of…
- LLegitQuestLegitQuest combines an Indian legal research database with tools that retrieve authorities, break down judgments and display later case treatment. iSearch finds relevant material, iDRAF…
- OOne LEGALEOne LEGALE is Wolters Kluwer Italia's online legal research platform for Italian lawyers, built on legislation, commented case law, doctrine, journals and practice tools from the group's…
- SSapient-IASapient-IA is the generative AI system of the Italian legal publisher Lefebvre Giuffrè, built to work only from the publisher's own editorial sources.
- OOtto Schmidt AnswersOtto Schmidt Answers is the AI research assistant of Verlag Dr. Otto Schmidt, the Cologne legal and tax publisher, built into its Otto Schmidt online databases.
Capability grades
All 15 axes, graded from public sources on the date shown. Hover a grade to see what the letter means on that axis.
AI Centrality
How much of the product is actually AI. Whether the machine learning is the mechanism the buyer is paying for or a feature layered onto conventional software, and whether the vendor is specific about which is which.
Specialist models perform the case analysis, citation review, and drafting. The Petition Reviewer identifies the practice area, the type of filing, and which party the attorney represents. It evaluates the relevant passages and checks cited law. A paid agent can extract a timeline from multiple records and suggest legal actions based on the documents. The matter workspace keeps the fact memorandum, draft, and discussion together, and its dashboard tracks work across agents.
Those records are created by the analysis tools. The product is sold as an AI workspace for working through evidence and developing a filing.
Citation Accuracy and Hallucination Disclosure
Whether the vendor publishes measured accuracy on citations and assertions, grounds output to primary sources, and says plainly what its system does when it does not know. Legal has a documented public record of fabricated citations reaching filed briefs, so an untested claim of accuracy is not evidence.
The Petition Reviewer checks cited precedents against LawX's own collection, marks each as current and links it to the deciding court. It marks each cited statute as adequate or needing verification and can suggest further authorities. In the integrated workspace, suggested actions link to the STJ, STF or state court precedents behind them, with a note on fit. The analysis separates a fact confirmed by a document from one that is only alleged.
Specialists update the collection daily. LawX shares no test set, measured citation error rate or comparison with lawyers checking the same filings.
Autonomy and Oversight Model
What the system decides on its own, what a lawyer must approve, and whether the vendor documents where the review point sits. A tool that drafts under review and a tool that files without one are different products and different risks.
A lawyer starts the analysis, selects an area and agent, and supplies the documents. The model can proceed through reading, sorting, and matching legal actions before returning a structured analysis. Possible actions include a confidence indication and questions that remain unresolved. A draft begins after the lawyer chooses a strategy. The lawyer can change or remove arguments, edit the draft, and work through missing information before exporting it.
Free petition review returns annotated passages and citation checks to a reviewer instead of filing the document. The firm retains the decision on the legal position and the text sent to the court.
Operational and Outcome Evidence
Named, dated evidence that the product works in production at real firms or legal departments. Case studies with figures and identified customers count. Unattributed testimonials and launch announcements do not.
A Brazilian social security example describes the analysis weighing 36 possible action types and suggesting three from the case material. A demonstration organizes ten supplied documents into an analysis in one minute and 17 seconds, and the LawX site claims 31 documents analyzed in three minutes. ADVBOX chief executive Eduardo Koetz introduced the reviewer and the paid agents in November 2026. The site carries a line of trust from more than 50,000 lawyers, with no count of firms on the current release and no named firm reporting results before and after LawX.
Privilege and Confidentiality Posture
How client confidences are handled: attorney client privilege and work product treatment, segregation of one client matter from another, whether client data trains any model, and what the vendor commits to in writing rather than in marketing.
The product describes a firm boundary around a case analysis. Its dashboard uses access levels similar to those in ADVBOX. The ADVBOX agreement establishes customer ownership of raw data within its licensed software and a mutual confidentiality duty. LawX's free Petition Reviewer says the filing is not saved. Review, firm history, and imported documents have their own treatment, with unfinished case work and drafts available in the integrated history.
A commitment covering training on case materials, legal privilege, and the retention of excerpts or output in the current LawX service is not specified.
UPL and Professional Responsibility Posture
Whether the vendor is clear that it supplies a tool rather than legal advice, who its audience is, and how it addresses unauthorized practice of law, competence and supervision duties, and jurisdiction limits. ABA Formal Opinion 512 is the reference point. Where the advice line is not the duty a product raises, the axis is read through the nearest professional duty it does raise: judicial conduct rules and the reviewing duty for products sold only to courts, and the duty to bill for time actually spent for products that draft time entries.
The software is designed for Brazilian legal work, with agents matched to subject areas and procedure. The lawyer controls which legal position and requests appear in a filing. Suggested actions identify relevant law and can be edited or removed before drafting. The Petition Reviewer shows which language or authorities need attention before filing. The earlier LawX service terms reserve responsibility for the use of output to the customer and describe the service as a technology tool.
The current product leaves professional decisions with the lawyer and does not map these review functions to specific duties under Brazilian bar guidance.
AI Governance and Bias Disclosure
Published governance over model behavior: who owns it inside the vendor, what is tested before release, and what is disclosed about disparate output across matter types, parties, or populations.
LawX splits its work into agents for Brazilian practice areas, each organizing the case and suggesting arguments from the law, procedure and standard filings of that area. The vendor describes a design choice that favors individuals and small businesses over institutional parties. Specialists curate the precedent collection, and the lawyer decides which arguments enter the petition. Those choices set whose position an agent is built to support and where human review comes in.
How that behavior varies by party type, practice area or court has not been tested in any result LawX shares, and no person is named as accountable for model conduct.
AI Safety and Data Stewardship
Retention, deletion, access control, and what happens to prompts and documents after they are processed. Whether the vendor states its subprocessors and its incident practice, or leaves the buyer to assume.
The free reviewer states that uploaded petitions are not kept. Integrated cases retain drafts and activity in the firm's ADVBOX history, where access follows the user's permissions, and a processed case takes no new documents, so added evidence starts a new case. ADVBOX's privacy policy describes encrypted transmission and notice when a personal data incident may pose a risk, and its software terms cover backup restoration and the handling of customer data after the contract ends.
LawX relies on those controls in the integration. How long prompts and outputs are stored, and which AI processors handle the standalone reviewer and the paid agents, are left open.
AI Liability and Recourse
What the vendor stands behind contractually when its output is wrong. Indemnities, caps, carve outs, insurance, and whether any of it is published or only reachable through a negotiated agreement.
Improvement notes and suggested authorities in a petition review are delivered for an attorney to consider before filing. The separately dated LawX terms disclaim loss from using generated output. The general ADVBOX software agreement applies its own fee based limit and provides a separate remedy for certain failures in transmitting court notices. Notice delivery is a different service from drafting or checking a petition.
No product clause extends that remedy to the reviewer's citation decisions or to an incorrect legal argument suggested by a paid LawX agent. Decisions about filing remain with the lawyer.
Practice Systems Integration Depth
How deeply the product reaches into the systems legal work already lives in: document management such as iManage and NetDocuments, Word and Outlook, contract lifecycle management, matter management, e-billing, and court filing systems.
Inside ADVBOX, a lawyer opens Legal AI Agents under Models and AI, selects the client by name, CPF or CNPJ, and the case's documents load into the analysis. Files can also come from a computer upload or a Google Drive link, and the privacy terms for that connector describe an authorization step for selected files. The social security workflow can bring in WhatsApp content through Atende Direto or Chat Jurídico, including photos and audio, which the agent transcribes into the timeline.
Export moves the fact memorandum to Markdown and the petition draft to Word. An ADVBOX dashboard shows the resulting activity by permission level.
Deployment Model and Data Residency
Where the software runs and where the data sits. Multi tenant cloud, single tenant, private deployment, on premises, and whether region of residence is a published option or an enterprise conversation.
The case workspace is accessed through an online account or from the integrated ADVBOX legal work system. The product describes case analysis as restricted to the firm and describes processing in Brazil. The earlier LawX integration policy distinguishes services such as Google and Mailchimp, whose servers can be outside Brazil. The firm boundary governs access to a firm's case history and analytics. Those arrangements do not specify where the underlying model handles each step, whether a different region is available, or where all external services store their data. The public materials describe a cloud workflow and partial location information.
Security Certifications and Trust Center
Independent attestation a buyer can pull without a sales call: SOC 2, ISO 27001, penetration test summaries, a trust center with current reports and named scope rather than a badge image.
Access to the integrated dashboard depends on the user's permission level. The current LawX site describes the same security infrastructure as ADVBOX and ties its personal information treatment to Brazil's LGPD. The ADVBOX privacy policy describes encrypted traffic and a commitment to notify affected users of qualifying incidents. The product controls restrict each analysis to the firm that created it. The available security information discusses privacy and operating practices, without an independent attestation naming the LawX reviewer or agents, an audit period, or a trust report for the current product.
Model Supply Chain Disclosure
Which models sit underneath, whose they are, where they run, and whether the vendor commits to telling customers when that changes. A legal buyer inherits every dependency it cannot see.
Practice area agents combine case facts, selected legal material and drafting structures for each area. They generate a chronology and proposed arguments, and the Petition Reviewer matches an uploaded filing to the right agent. The daily updated precedent collection supplies the legal material those workflows use. The earlier platform terms describe generating content through prompts and APIs, and the LawX site lists a native Claude integration as coming soon.
Which foundation models run each agent, who supplies inference, where those models run and how customers hear of a model change all go unnamed.
Commercial Transparency
Whether a buyer can learn what this costs without entering a sales process: published rates, the unit being charged, what sits behind an enterprise tier, and what implementation adds.
The paid offer is priced per practice agent at R$200 a month, so a firm can subscribe to social security, labor, family or another area without buying every practice. The Petition Reviewer is free for up to five reviews a day after email confirmation. Access to the paid agents is rolling out by firm, and the LawX site runs a waitlist with no free trial of the agents. ADVBOX practice management is sold on its own plans and can supply records to LawX where the firm connects it. Setup charges, annual rates and volume discounts for the agents are not listed.
Firm and Practice Coverage
Who the product is actually built for. AmLaw, midlaw, small firm and solo, in house departments, government and courts, and which practice areas are supported rather than merely claimed.
The practice library serves Brazilian lawyers at small firms, larger teams that file the same petitions repeatedly, and legal departments. The free reviewer lists 18 agents, and the LawX site claims 17 practice areas, naming social security, labor, family, welfare assistance, digital, tax, administrative, environmental, banking, consumer, criminal, real estate, business, health and civil work. Every agent shares eight base petition types.
The social security agent handles contribution histories and expert reports. The work rests on Brazilian law and ties its research to the Brazilian courts.
5 public documents
The public pages on file for LawX Brazil, with the recorded signals each one supports and the date it was last read. Open any of them and check the reading against the record.
-
advbox.com.br/blog/lawx3 signals
Primary Law Corpus Provenance, Good Law Verification, Billing and Fee Posture
Read Nov 20, 2026
-
lawx.ai2 signals
Ethical Walls and Matter Segregation, Refusal and Uncertainty Behavior
Read Nov 20, 2026
-
base.lawx.ai/revisor1 signal
Prompt and Output Retention
Read Nov 20, 2026
-
Court Disclosure Support
Read Nov 20, 2026
-
Third Party Request and Subpoena Notice
Read Nov 20, 2026
R$200 per agent per month
- A paid LawX agent costs R$200 a month.
- A firm can pay for agents in the practice areas it uses.
- The Petition Reviewer offers up to five free reviews each day after email confirmation.
- Paid access is rolling out to firms in stages.
The rate is per practice agent, such as the social security agent. The free Petition Reviewer is separate and limited to five reviews a day. ADVBOX case management has its own subscription.
Implementation: The standalone Petition Reviewer runs through a LawX account. The paid agents are not shown with a setup charge.
Note: Access to paid LawX features is being released in stages. The public reviewer verifies email before the free allowance can be used.
Legal Signals
What each signal meansA signal records what public sources say on the date shown. It is not a grade and it is not a recommendation. Where a signal reads Not addressed, it means the index did not locate the material in public sources on that date, which is a statement about disclosure rather than about the product.
Client Data in Training
Can material a lawyer puts into this product be used to train a model?
A published agreement or policy exists and none of it addresses the question either way, or the document that would answer it could not be read and the summary names the retrieval limit. The summary states which shape the silence takes: an improvement right granted that never names training, or no improvement right granted at all.
The terms and privacy materials for LawX describe service improvement and a firm's control of sensitive information, and they leave model training on submitted case material unaddressed. The ADVBOX agreement protects customer data in the integrated workspace without a training term either way. The reviewer's statement that a petition is not kept is the only data commitment specific to LawX.
Prompt and Output Retention
How long does the product keep what a lawyer typed, and can that be set to zero?
Retention is acknowledged in public materials with no stated period.
The public Petition Reviewer says the submitted petition is not kept. Drafts from the paid case workflow stay in the ADVBOX case history, and the prompts and AI responses behind them carry no stated lifetime.
Ethical Walls and Matter Segregation
Does retrieval respect the firm’s ethical walls, or can the model read across them?
Segregation is asserted in public materials with no published detail on how it is enforced.
Each analysis is restricted to the firm that produced it, and ADVBOX roles apply to the dashboard. Rules for walling off a particular matter inside a firm, or for how excerpts are handled when they reach the model, are not described.
Third Party Request and Subpoena Notice
If someone subpoenas the vendor for a firm’s data, does the firm hear about it first?
Published terms or policy address disclosure to authorities or in response to legal process, and no commitment or reservation regarding customer notice is located anywhere. The vendor has told the customer that data can leave and has said nothing about whether the customer hears of it.
The earlier LawX privacy terms permit disclosure of selected Google Drive information when needed to run the feature or when required by law. Neither those terms nor the ADVBOX policies that now govern the integrated workflow promise to tell the customer about a legal request for its data.
Primary Law Corpus Provenance
Where does the law in this product come from, and does the vendor have the right to use it?
Sources are identified without stating the license or rights basis.
A team of specialists curates the LawX precedent collection and updates it daily, and suggested actions link to STJ, STF and state court decisions. The collection is described by its courts and normative weight, with no stated rights basis or source list for the decisions it holds.
Good Law Verification
Does the product tell you when the authority it just cited has been overruled?
The vendor computes and surfaces subsequent history itself, with the method described.
The reviewer compares each cited precedent with LawX's own collection, marks whether it is current and links to the deciding court, and it may suggest further precedents. Each cited statute is marked adequate or needing verification. The treatment comes from LawX's curated collection rather than an outside citator.
Refusal and Uncertainty Behavior
What does the product do when the answer is not in the corpus?
The product exposes a confidence or grounding score without an explicit abstention path.
The analysis gives each possible legal action a confidence level, labels facts confirmed or alleged, and flags missing or contradictory support. The reviewer marks cited statutes that need verification. LawX does not say it declines to answer when no source supports a proposition.
Fabricated Citation Record
Does a public court record exist addressing fabricated or hallucinated legal citations in output from this product?
No court order, opinion or disciplinary record addressing fabricated or hallucinated legal citations produced by this product has been located as of the date shown. This is a statement about the public record on that one subject, not a finding about the product, and this signal is not a litigation history.
The AI Hallucination Cases database kept by Damien Charlotin has no entry naming the Brazilian LawX service. LawX drafts and reviews filings for Brazilian courts and checks cited precedents against its own collection.
Bar Guidance Alignment
Has the vendor engaged in public with the ethics opinions its buyers are bound by?
No located public material engages with bar or ethics guidance.
The lawyer chooses the strategy and the draft language, and ADVBOX's help guide tells the responsible professional to verify AI output before use. The reviewer and the agents are not mapped to the Brazilian bar's guidance on generative AI, and no step in the product refers to the OAB's recommendations.
Billing and Fee Posture
Does the vendor address what happens to the bill when the work takes an hour instead of six?
Public materials claim time savings without addressing billing or disclosure, and the product sits inside a fee relationship between a lawyer and a client where those savings would change the bill.
The product sells on speed, with ten documents analyzed in one minute and 17 seconds, 31 documents in three minutes and petition reviews in under a minute. The ADVBOX dashboard shows work by category and period. No feature attributes AI assisted time or cost on a client invoice.
Outside Counsel Guideline Readiness
Can a firm get this vendor through a client’s AI clause without a bespoke negotiation?
No located public material supports a client side disclosure obligation.
The service describes a firm level access boundary, LGPD compliance and the same security structure as ADVBOX. It offers no current list of the providers behind its AI workflows and no disclosure package a firm could hand to a client that sets outside counsel guidelines.
Court Disclosure Support
If a judge’s standing order requires an AI disclosure, can the product produce one?
Some elements of the record are available, short of a document level export.
Drafts stay in the integrated case history. The fact analysis gives a source for each factual reference, labels each fact confirmed or alleged, names the agent that ran it, and the reviewer links to the authorities it checks. Those pieces sit apart, with no single export prepared for a court's AI disclosure.