BriefCatch vs Clearbrief: how they compare in 2026
BriefCatch and Clearbrief both sit inside Word and check whether the authorities in a brief say what the brief claims, and both count courts among their users. BriefCatch is stronger on its published commercial and data terms. It publishes seat prices from $60 a month with a stated credit cost per brief, and a contract clause says it retains no customer content. AI is switched off by default with an administrator control, and it offers an intellectual property indemnity. Its RealityCheck confirms a cited case across three case law sources and tells the user to check by hand any it cannot confirm. Clearbrief's counterweight is how it checks and what it leaves behind. Its checker runs on classic machine learning rather than a generative model, so the checker cannot invent anything. Its Cite Check Report is a downloadable audit trail the signing attorney can keep, and a judge can open its filings without a subscription. Neither checks whether a cited case is still good law.
Which one fits
Choose BriefCatch if
- You need nothing left on the vendor's servers. BriefCatch's master terms state that it does not store or retain customer content, document text is processed in memory and cleared, and AI is off by default, so in its default setting the document never leaves your own Word environment.
- You want quotations and pincites checked, not just case names. BriefCatch's RealityCheck confirms that cited cases and federal statutes exist, then checks whether quoted language appears in the opinion, whether pincites are right and whether a dissent is presented as a holding, and flags any case it cannot confirm across three sources for manual review.
- You want writing edits and verification at a known price. BriefCatch publishes Individual seats at $60 and Pro at $75 per month billed annually, with 650 or 700 AI credits a seat and about 70 credits for a typical brief, plus more than 12,000 rule sets for clarity and concision.
Choose Clearbrief if
- You want the checker itself unable to invent anything. Clearbrief runs its verification on classic machine learning and natural language processing rather than a generative model, scoring how well each sentence is supported by the page it cites and flagging weak support, missing sources and formatting errors.
- You want a record the signing partner can keep. Clearbrief's downloadable Cite Check Report lists every factual and legal citation with links to each flagged item, and it publishes hyperlinked filings that a judge or client can open without a Clearbrief subscription.
- You want documents kept in your own cloud and connected to your systems. Clearbrief's enterprise Bring Your Own Storage keeps uploads in your own cloud environment, and it names integrations with Relativity, iManage, NetDocuments and Clio, and with LexisNexis, Fastcase and vLex for authority.
At a glance
All 15 axes, side by side
The same grid applied to every vendor in the index, graded from public sources. 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.
AI is off by default, and with it off the suite runs on its rule engine: more than 12,000 rule sets for grammar, clarity, style and structure, plus scoring and reports, which BriefCatch says the AI switch does not affect. The AI adds three things. BriefChat is a writing adviser. The AI half of CiteCheck refines rule based citation fixes. The second layer of RealityCheck judges whether quoted language is in the opinion and whether the authority supports the point. RealityCheck's first layer, confirming that a case or federal statute exists, is deterministic. Since September 2026 RealityCheck is on every plan and metered in AI credits.
Citation detection across a document, the patented scoring of how well a sentence is supported by the page it cites, and concept search across uploaded discovery all run on machine learning and natural language processing. The vendor keeps generative models out of the checking function, saying classic AI means the checker cannot introduce hallucinations of its own, and ships separate generative features for drafting. Without the models, what remains is a Word toolbar.
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.
Deterministic lookups first confirm that a cited case or federal statute exists, matching reporter volume, court identifiers and case names against authoritative legal databases. AI models then read the opinion to judge whether the quoted language is in it and whether it supports the point. Results come back as correct, needs attention or cannot verify, each pointing to the authority concerned. No precision or recall figure, test set, false positive or false negative rate, or third party evaluation is published for RealityCheck, CiteCheck or BriefChat. Clause 8.2 of the Master Terms says output is not guaranteed correct, complete or current. A reported test on a Fifth Circuit brief in Fletcher v. Experian comes from trade press, not the vendor.
Citations are detected in the Word document and matched against case law databases: LexisNexis where the user holds a subscription, and Fastcase and vLex where they do not, with Westlaw citations mapped across to those databases. Where a source cannot be displayed automatically, the product still reports that the citation is verifiable and accepts a manual upload. Output links every citation to the specific page and text the reader can open. Because checking runs on classic AI rather than generative models, the vendor says the checker carries no risk of adding hallucinations of its own. No false positive or false negative rate for citation detection, accuracy figure for the support score, test set or independent evaluation is published.
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.
Each suggestion is a card the user accepts or rejects, so nothing changes a document without a person acting. RealityCheck sorts results into confirmed, needs attention and cannot verify, so unresolved cases stay visible. Clause 8.2 of the Master Terms requires all output to be reviewed before use, and clause 8.5 bars relying on output as legal advice. The system never acts on its own. Nothing says what happens after an output is found wrong, or when CiteCheck's AI overrides or defers to the rule engine.
The vendor states that source linked outputs keep humans in control and the evidence in view. The Cite Check Report is built for the partner signing a filing, listing every citation with hyperlinks so each flagged issue can be reviewed in context. The founder frames the problem as partners being sanctioned for citation errors they did not personally make. No threshold is published for when a semantic score counts as low, nothing says whether that threshold is configurable, and no commitment covers what happens when the checker misses something.
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.
Skadden, Wilson Sonsini, Nixon Peabody, Kellogg Hansen and Bartlit Beck appear as customers on BriefCatch's pricing page. It states more than 25,000 lawyers and 40 of the Am Law 200 as users, plus adoption by state and federal courts, with pages for courts and government. It shows a 4.9 rating on Microsoft AppSource. The one figure, more than 50 percent of editing time saved, is not tied to a customer, dated or explained. The site has a Case Studies section.
The vendor states the product is used by courts, Am Law firms, government agencies, arbitrators and legal teams handling high stakes matters, and an attributed testimonial describes the LexisNexis backed hallucination detection in practice. No named organization is published with figures and a date, and no case study or measured method appears on the vendor site, the blog or the sandbox site, or in trade coverage.
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.
Clause 7.2 of the Master Terms says Customer Content is not stored or retained, so there is nothing to return at termination, and BriefCatch's Trust Center says text is processed in memory only and promptly cleared. Training is excluded in express terms on several pages. Documents stay in the local Word environment unless the user turns on AI citation review or BriefChat, and BriefChat sees only prompts, never the document or email. Customer Content is the customer's confidential information under clause 5.1, and clause 5.2 confirms the customer owns it. Privilege and work product are not named, and no model provider is identified. The EULA says the products are not meant for confidential or personal data and puts privacy risk on the user, while the product FAQ says the product is safe for confidential client documents.
Bring Your Own Storage, an enterprise option, keeps all uploaded documents in the organization's own cloud storage, so the vendor does not hold custody of them. The vendor says the product is built for sensitive privileged legal work with enterprise grade security, privacy and access controls, and that work stays in the environment the team and IT already trust. A policy of not using client data for model training is reported in third party review material and is not stated on the vendor's own site. Retention and deletion terms for customers not using Bring Your Own Storage are not published. Attorney client privilege and work product describe the work rather than being addressed as a handling question, and no segregation model between users or matters is published.
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.
Clause 8.5 of the Master Terms bars the customer from relying on output as legal advice, and from telling anyone that output is legal advice. Clause 8.6 says the service gives no legal advice and creates no attorney client relationship. For ClientCatch, recipients of output must be told to consult a qualified attorney. Nothing covers supervision or competence, such as who may use the AI features or what review a supervising lawyer applies. No rule of professional conduct or bar authority is named, and no jurisdiction limit is stated for a product whose citation checks are US specific.
Clearbrief publishes material on how large firms put responsible AI into practice through citation checking policies, describes the courts' expectation that every sentence carries a verifiable citation in a standard format, and frames the risk as sanctions and reputational damage to the signing partner, covering candor to the court and supervision of others' work. The audience is professional throughout: litigators, in house teams, courts, arbitrators and government agencies, with no consumer product. Competence is not addressed as such, no statement on the advice line is published, and no jurisdiction limits are stated beyond references to Bluebook and local style rules.
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.
AI features are off by default, and administrators can switch AI on or off across the whole organization while the rule engine keeps working. A dedicated AI Disclosure page explains it. BriefCatch states it sends no more data than needed for a suggestion, explains what its models were and were not trained on, and says the AI switch affects only BriefChat and citation suggestions. No person or committee is named as accountable for model behavior, no testing before release is described, and no evaluation result is published. Bias is not addressed, though the product scores writing against the styles of named judges and advocates.
The vendor publishes on how firms should put responsible AI into practice through citation checking policies, and keeps generative models out of the checking path for a stated reason. No AI governance framework, AI management certification such as ISO 42001, named owner of model governance or testing regime before release is published. Nothing addresses uneven output across matter types, parties or populations, or whether the support score behaves differently across drafting styles or document types.
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.
Retention is zero and in the contract: clause 7.2 of the Master Terms says Customer Content is not stored or retained, and BriefCatch's Trust Center says document text is processed in memory only and promptly cleared. Encryption is AES-256 at rest and TLS 1.3 in transit. Access controls include two factor and Google authentication on every plan, and single sign on, SCIM provisioning and role based access at enterprise. Backups, redundancy and a tested disaster recovery plan are described, with staff confidentiality agreements and annual training. RealityCheck also scans documents for text, such as white on white type and zero width characters. No subprocessor list beyond Azure for hosting and no breach notice commitment to customers is published.
Bring Your Own Storage lets an organization keep uploaded documents in its own cloud. The vendor says it works with customer security and IT teams through procurement and review, and describes enterprise grade security, privacy and access controls in general terms. No retention period, deletion control, encryption detail, named subprocessor list, or incident or breach notification practice is published. Published filings are hosted in the vendor's cloud for judges and clients to read, a second data location with no published terms attached.
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.
Clause 8.1 of the Master Terms commits to reasonable, industry standard efforts to minimize errors and interruptions, workmanlike performance, and advance notice of scheduled disruption. Clause 9 indemnifies the customer against third party claims that the service infringes a US patent or copyright or misappropriates a trade secret, with BriefCatch running the defense and replacing, licensing or refunding if needed. Liability is capped at twelve months of fees, and the cap does not apply to gross negligence, willful misconduct or breach of confidentiality. A service level agreement is incorporated, with 99.9 percent uptime on the enterprise tier. Clauses 8.2 and 8.3 disclaim responsibility for output and anything based on it, no insurance position is published, and the indemnity does not reach a claim from a wrong verification.
Clearbrief publishes no indemnity, liability cap, carve out, warranty on output or insurance position, and no customer terms of service. The product is bought to keep sanctionable citation errors out of filings, and nothing published covers what happens if it misses one.
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.
BriefCatch runs as a Microsoft Word add in, an Outlook add in at Pro and above, and a browser editor, and works on Word files in Google Drive and OneDrive. It is distributed through Microsoft AppSource with a Microsoft 365 App Certification listing. Identity options are documented by tier: two factor and Google sign in on every plan, and single sign on with SCIM provisioning at enterprise. Learning content plugs into a customer's learning system over SCORM. Document management connections are an enterprise feature described only generically, with no system named, no direction of data flow and no setup requirements.
Four document repository integrations are named: Relativity, iManage, NetDocuments and Clio, letting a firm keep its existing secure infrastructure and add the checking layer on top. Delivery is a native Microsoft Word add-in, where drafting and signing happen. Research integrations are named with their commercial terms: LexisNexis where the customer holds a subscription, and Fastcase and vLex where none is required, with Westlaw citations mapped across so they stay checkable. No integrations index page or documentation for each integration describes what it moves, in which direction, or what an administrator configures.
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.
BriefCatch's Trust Center says the service runs on Azure in United States locations only. It is delivered as a browser app and as Word and Outlook add ins, through Microsoft AppSource or direct install, with a security and privacy white paper for IT review on request. With AI off, the default, documents never leave the local Word environment. They are sent only when the user turns on AI citation review or BriefChat. Nothing says whether accounts share an environment, and no private or on premises option is described.
Two models are published: the standard cloud service, and an enterprise Bring Your Own Storage option under which uploaded documents stay in the organization's own cloud storage, which the vendor presents as giving more control over data residency and confidentiality. With customer held storage, the customer keeps custody. No hosting provider for the standard service, named regions, tenancy model or statement of where processing happens as distinct from storage is published. Processing still takes place when the customer holds the storage, and its location is not stated.
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.
A public Trust Center, reachable without a form, states BriefCatch's SOC 2 certification, AES-256 and TLS 1.3 encryption, a tested disaster recovery plan, firewalls, intrusion detection, continuous logging, staff confidentiality agreements and annual training. The pricing table lists SOC 2 Type II on every tier. No report, certificate, audit period or auditor is available or named, and no scope says which systems are covered. BriefCatch's pricing page lists SOC 2 Type II, while a BriefCatch 3 support FAQ that is still live says SOC 2 Type 1. Type 1 tests design at a point in time, and Type II tests operation over a period. The ISO 27001, PCI DSS Level 1 and SOC 2 compliance on the Trust Center belongs to Azure, not BriefCatch.
The vendor's blog states SOC 2 Type 2 certification, describing it as showing that it meets rigorous data protection standards and as helping a lawyer meet the duty to protect client confidentiality through verified security controls. The same material says Bring Your Own Storage lets a firm control where client data is stored to meet jurisdictional requirements or firm retention policies. No coverage period, audit scope, report date, named auditing firm, trust portal or published request route is published. Security diligence runs through the customer's security and IT teams during procurement.
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.
Models are trained only on proprietary, licensed or public non customer data, which the pricing FAQ calls real filings, judicial opinions and decades of legal writing expertise, with BriefChat trained on the founder's published writing. RealityCheck's first layer checks against authoritative legal databases. No model, version or provider is named, and nothing says which models BriefCatch built and which it licenses. BriefCatch's FAQ says documents never improve third party AI services, without naming any third party. Azure is named as the host. Trade press reports that a supplier called Counsel Stack provides RealityCheck's citation infrastructure. It is not named anywhere on BriefCatch's own site, including the Master Terms and Trust Center.
The checking path runs on classic machine learning and natural language processing with a patented scoring system, not generative models, so it does not depend on a third party foundation model. Generative features exist separately for drafting. No model or provider behind the generative features is named, and no statement of where models run, subprocessor list or commitment to notify customers of supply chain changes is published.
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.
Individual is $60 per seat per month billed annually for one seat, and Pro is $75 per seat per month billed annually with a five seat minimum. Enterprise is custom above twenty seats. AI use is metered in credits: 650 a month on Individual and 700 on Pro. BriefCatch publishes a conversion, about 70 credits for a brief under 30 pages with fewer than 100 citations. Credits reset monthly without rollover, extra packs last a year, annual billing is discounted, and plan changes keep credits and settings. A full feature table compares tiers, and a seven day trial needs no credit card. Enterprise pricing is not published, and the get started page still shows a Solo plan at $599 a year that does not match the pricing page.
No pricing page, rate, unit of charge or tier structure is published, and every commercial path ends in a demo or contact request. Third party review material states that pricing is custom quoted. A recipient of a hyperlinked filing, including a judge or a client, does not need a Clearbrief subscription to open it.
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.
BriefCatch has pages for law firms, government and courts, and sells to courts to check filings they receive, stating adoption by state and federal courts. Tiers run from one seat to a five seat practice group to twenty or more, and subscriptions are offered to solo lawyers, students, educators, firms, universities, corporations and agencies. Named customers include Skadden, Wilson Sonsini, Nixon Peabody, Kellogg Hansen and Bartlit Beck. No practice area or jurisdiction is stated, though citation checking works on US reporters and federal statutes and the writing rules come from US practice. Nothing says which document types beyond briefs, memos and emails are supported.
The stated users are litigators, in house legal teams, courts, arbitrators and government agencies, so the adopters include the tribunal as well as the advocate. Practice focus is litigation documents heavy with facts. Named uses cover briefs, tables of authorities, exhibit management, investigation reports for in house compliance teams, and real time trial support including cross examination outlines. No firm size segmentation or jurisdictional coverage is stated beyond references to Bluebook and local style rules, and no document type or practice area is named as out of scope.
The 12 legal signals, side by side
Recorded rather than graded. These are the questions a practitioner has to answer before a tool touches a client matter, and the answers are taken from public material only.
Client Data in Training
Can material a lawyer puts into this product be used to train a model?
BriefCatch excludes training on customer content in express terms on four pages. It does not use document text to train models, develop AI features or improve system behavior. Models are trained only on proprietary, licensed or public non customer data, documents never train shared models or improve third party AI services, and nothing written is used to update the software. Documents stay local unless AI is turned on, BriefChat sees only prompts, and clause 7.2 of the Master Terms says Customer Content is not stored.
No matching training term is in the agreement. Clause 5.6 lets BriefCatch analyze data about use of the service, including information concerning Customer Content, to improve the service and share it in aggregate or deidentified form, without naming training.
Third party review material summarizing Clearbrief's security posture reports an explicit policy of not using client data for model training. The vendor site, blog and sandbox site do not state the policy in the vendor's own words. Two vendor facts fit it without stating it: the Bring Your Own Storage option keeps uploaded documents in the customer's own cloud, and the verification path avoids generative models.
Prompt and Output Retention
How long does the product keep what a lawyer typed, and can that be set to zero?
Clause 7.2 of the Master Terms says BriefCatch does not store or retain Customer Content, so there is none to return at termination. BriefCatch's Trust Center says text is processed in memory only and promptly cleared. Documents do not leave the local Word environment unless the user turns on AI citation review or BriefChat, and BriefChat sees only prompts. Clause 5.6 lets BriefCatch keep data about use and performance of the service, including information concerning Customer Content, which is usage data rather than the documents.
No published material states how long uploaded documents, analyses or generated reports are retained, whether a customer controls the window, or whether deletion is available. Enterprise customers on Bring Your Own Storage keep uploaded documents in their own cloud and set retention by controlling the store; that option is not the default. Filings published as hyperlinked versions for judges and clients are hosted in the vendor's cloud, with no published retention terms for that store.
Ethical Walls and Matter Segregation
Does retrieval respect the firm’s ethical walls, or can the model read across them?
Access runs on BriefCatch's own model, set out by tier: two factor and Google authentication on every plan, and single sign on, SCIM provisioning and role based access over features and settings at enterprise. Administrators can switch AI features on or off for the organization. BriefCatch keeps no material: documents are processed in memory and cleared, nothing is stored, and by default the text never leaves the user's Word environment.
Nothing says whether roles can limit which users may turn on AI processing for which documents, rather than switching it for everyone.
No segregation between users or matters and no conflicts or ethical wall feature is described. The vendor refers to access controls that IT and security teams can trust without describing them. With Bring Your Own Storage, an enterprise customer's documents sit in a store it already governs, so its own controls apply to documents at rest. Nothing published says whether retrieval and concept search inside the product respect those controls for each user at query time, or whether iManage and NetDocuments permissions carry over into the product.
Third Party Request and Subpoena Notice
If someone subpoenas the vendor for a firm’s data, does the firm hear about it first?
Clause 5.1 of the Master Terms protects each party's Proprietary Information, including Customer Content, then excepts information required to be disclosed by law. No commitment to give notice, a chance to object or seek a protective order, or to narrow a demand appears, and no transparency report exists. Clause 7.2 says Customer Content is not stored, and BriefCatch's Trust Center says text is cleared from memory, so a demand would find account and usage records rather than documents.
No published customer terms or privacy policy commits to notice when a government or law enforcement body requests customer data, and no transparency report is published. Under Bring Your Own Storage, documents sit in a store the customer controls; the vendor does not say how that affects third party requests.
Primary Law Corpus Provenance
Where does the law in this product come from, and does the vendor have the right to use it?
The pricing FAQ says the product is trained on real filings, judicial opinions and decades of legal writing expertise rather than scraped internet data. The support documentation says all models are trained only on proprietary, licensed or publicly available non customer data, and no user content is used for training. BriefChat is trained on the published writing of founder Ross Guberman, the company's own material, and the editing rules draw on judicial and advocacy writing.
No individual database, publisher, reporter series or licensor is named, and the legal databases RealityCheck checks against are described only as authoritative.
The product holds no primary law corpus of its own. It displays cited authority through named third party databases: LexisNexis where the customer holds a subscription, and Fastcase and vLex where no subscription is required. Westlaw citations are mapped across to those databases so they can still be checked. The license or rights basis for access to those databases is not published, and completeness and update frequency for the underlying case law sit with the database owners.
Good Law Verification
Does the product tell you when the authority it just cited has been overruled?
BriefCatch does not check whether authority is still good law. Its marketing contrasts RealityCheck with a traditional citator: RealityCheck asks whether the authority says what the brief claims, not whether it has been overruled. It verifies that a case or federal statute exists, that quoted language is in the opinion, that pincites are right, that a case is not cited for a point it rejects, that a concurrence or dissent is not presented as the holding, and that signals such as But see are used correctly. None of that tells a user an authority was later overruled, distinguished or superseded.
The product confirms a cited case exists and is not fabricated, and scores how well the cited page supports the sentence's proposition, flagging low scores where the source may not support the assertion. No statement says the product checks whether a case has been overruled, distinguished or otherwise treated, and no commercial citator license for treatment signals is named. Third party comparison material refers to citation validity tools in general terms without settling the point.
Refusal and Uncertainty Behavior
What does the product do when the answer is not in the corpus?
RealityCheck tries to confirm a case across three case law sources, and when it cannot, it says so and tells the user to verify that case manually. Results are sorted into correct, needs attention and cannot verify, so unresolved cases show as their own group rather than passing silently. This is described in launch material and support documentation, without a published evaluation, and nothing says how often the cannot verify state occurs. Nothing addresses uncertainty in BriefChat or CiteCheck's AI.
The product evaluates assertions a lawyer has already written rather than answering legal questions from a corpus. Where a cited source cannot be displayed automatically because it belongs to another database, the product reports that the citation is verifiable rather than silently passing or failing it, and invites a manual upload. No statement covers what the product does when it cannot ground an answer.
Fabricated Citation Record
Does a public court record exist addressing fabricated or hallucinated legal citations in output from this product?
No decision, sanction or referral in reporting or trackers of court decisions on AI fabricated citations names BriefCatch, RealityCheck, CiteCheck or BriefChat. BriefCatch sells tools to catch this problem and cites Damien Charlotin's database of more than a thousand AI hallucination decisions. Neither BriefChat nor CiteCheck's AI has been named in a filing with a fabricated citation.
The AI Hallucination Cases database kept by Damien Charlotin holds no entry naming Clearbrief. The product checks citations in drafts a lawyer has already written rather than generating legal text.
Bar Guidance Alignment
Has the vendor engaged in public with the ethics opinions its buyers are bound by?
BriefCatch's marketing describes courts sanctioning and sometimes disqualifying lawyers, cites a database of more than a thousand AI hallucination decisions, and sells the product as pre filing quality control that protects professional credibility. The Master Terms bar customers from treating or presenting output as legal advice. No bar association, rule of professional conduct, ethics opinion or court standing order is cited, in any jurisdiction, though the sanctions it describes rest on duties such as candor, competence and Rule 11 certification. Nothing says which jurisdiction's citation conventions the checks apply.
The vendor publishes on how large firms put responsible AI into practice through citation checking policies. It describes the courts' expectation that every sentence carries a citation the court can look up and verify, refers to Bluebook and local style rules, and frames the risk as partners being sanctioned and suffering reputational damage for citation errors they did not personally make. No ethics opinion, including ABA Formal Opinion 512, and no state bar guidance or standing court order on AI disclosure is named on the vendor site or blog.
Billing and Fee Posture
Does the vendor address what happens to the bill when the work takes an hour instead of six?
Nothing published addresses how AI assisted work is billed. BriefCatch sells to law firms that usually bill brief drafting by the hour, and claims more than 50 percent of editing time saved. No record per matter of AI assisted work, marking of machine suggestions for fee narratives, or fee or disclosure guidance is described. The AI credit system shows what the customer pays BriefCatch, with 650 or 700 credits per seat a month and about 70 credits per short brief. Nothing connects credit use to the client's invoice.
The published claim is that the product greatly reduces time spent on manual nonbillable tasks such as fact checking and building tables of authorities, work a firm does not bill for. No per matter record of AI assisted work for fee purposes and no guidance on billing, fee or client disclosure treatment is published.
Outside Counsel Guideline Readiness
Can a firm get this vendor through a client’s AI clause without a bespoke negotiation?
There is no subprocessor list, and the only third party named is Microsoft Azure, as host. No model or provider is named, and BriefCatch's FAQ says documents never improve third party AI services without naming them. Trade press says Counsel Stack supplies RealityCheck's citation infrastructure, but it appears nowhere in the Master Terms or Trust Center, and nothing says whether document content reaches it. The published Master Terms commit at clause 11.2 to a data processing agreement where the law requires one, and the Trust Center is public. Neither says who else touches the data.
The vendor says it works with customer security and IT teams through procurement and review, supplying diligence material in that process rather than as a published pack. Third party review material reports SOC 2 Type 2 certification and a no training policy. Bring Your Own Storage keeps client documents in the firm's own cloud. No subprocessor list, statement of which model providers see client content, data processing agreement, or consent or notification pack for clients is published on the vendor site, blog or sandbox site.
Court Disclosure Support
If a judge’s standing order requires an AI disclosure, can the product produce one?
Nothing published addresses disclosing BriefCatch's own AI involvement in a filing. Courts use RealityCheck to check briefs filed by others. Nothing shows a court what BriefChat drafted or which AI refined citation fixes a lawyer accepted. No record of AI involvement, marking of accepted AI suggestions, export, certification template or disclosure guidance is described. The Master Terms bar presenting output as legal advice, which limits what users say rather than supporting disclosure. AI is off by default.
The Cite Check Report is a downloadable record for each document, listing every factual and legal citation it contains and flagging missing sources, formatting errors and low support scores. Each item links back so the signing attorney can review it in context, and the vendor calls it an audit trail of citation review activity. Linked versions of filings can be opened by a judge or client without a subscription, so the sources relied on travel with the document to the tribunal.
The report covers sources and human review at document level and does not record a model, since the checking path runs on classic machine learning rather than a generative model.
The questions both sides leave open
Derived from the records above rather than written, so it cannot favor either vendor. Take these into both conversations and ask each side the same question.
- Good Law Verification
Product updates
The latest logged changes from each vendor, newest first, each linked to its source.
Clearbrief
-
Product / capability
Clearbrief's Word analyzer now recognizes public domain case citations from Illinois, Ohio, Pennsylvania and Louisiana and formats them correctly in its Table of Authorities.
Source
In summary
BriefCatch
BriefCatch, from LawCatch, Inc. of Washington DC, is a legal writing and authority verification suite that runs in Word, Outlook and a browser editor for lawyers, judges and courts. Its rule based engine applies more than 12,000 rule sets for clarity and persuasiveness. Its AI features, off by default, are CiteCheck, the BriefChat writing adviser and RealityCheck, which verifies that authorities exist and support what they are cited for. According to the AI Legal Index, BriefCatch publishes its pricing, from $60 per seat a month, and its terms state that no customer content is retained. It names Skadden and Wilson Sonsini among customers. No model provider or AI subprocessor is named.
Clearbrief
Clearbrief, founded in 2020 in Seattle by former assistant attorney general Jacqueline Schafer, is a citation and fact verification add in for Word. It is used by litigators and in house teams, and by courts, arbitrators and government agencies. It detects every citation to facts and law and scores how well each sentence is supported by the page it cites, using classic machine learning rather than a generative model. It produces a downloadable Cite Check Report as an audit trail. The AI Legal Index records a Bring Your Own Storage option and named integrations with Relativity, iManage, NetDocuments and Clio.
Questions buyers ask
BriefCatch vs Clearbrief: which is better for checking citations in a brief?
Both check whether authority supports what it is cited for. BriefCatch publishes its prices, retention terms and liability position. Clearbrief checks factual citations to the record as well, avoids generative models in its checker and produces a downloadable audit report. Firms that need a record for the signing partner have more from Clearbrief. From the AI Legal Index, based on each vendor's own published materials as of September 25, 2026. No vendor pays for placement.
Do BriefCatch or Clearbrief check whether a case is still good law?
No. BriefCatch states that RealityCheck asks whether an authority says what the brief claims, unlike a citator that tells a lawyer whether it is still good law. Clearbrief confirms a cited case exists and scores how well the cited page supports the sentence. A lawyer still needs a citator for overruled or superseded authority. From the AI Legal Index, based on each vendor's own published materials as of September 25, 2026. No vendor pays for placement.
Does BriefCatch keep my documents?
Its master terms state that it does not store or otherwise retain customer content, and its trust center says document text is processed in memory and promptly cleared. With AI off, which is the default, documents do not leave the local Word environment; turning on AI citation review or BriefChat sends text for processing. Its end user license separately states that the products are not intended for confidential or personal data. From the AI Legal Index, based on each vendor's own published materials as of September 25, 2026. No vendor pays for placement.
What is Clearbrief's Cite Check Report?
It is a downloadable report for each document that lists every factual and legal citation, flags missing sources, formatting errors and citations with weak support, and links to each flagged item so the signing attorney can review it in context. Clearbrief describes it as an audit trail of citation review. It does not record which model produced anything, since the checking runs on classic machine learning. From the AI Legal Index, based on each vendor's own published materials as of September 25, 2026. No vendor pays for placement.
Neither vendor paid for inclusion, placement or a grade, and neither reviewed this page before it published. Everything on it comes from public material on the dates shown. How the index grades.
Three points to weigh. BriefCatch's end user license says its products are not intended for processing confidential or personal data. Its master terms, by contrast, treat customer content as confidential and state that none is retained. BriefCatch's own pages state SOC 2 Type II in one place and Type 1 in another. Clearbrief's policy against training on client data appears in third party review material rather than on its own pages, and it publishes no customer terms. Neither vendor reviewed this page.