CaseMine vs Descrybe: how they compare in 2026

The models are the engine of everything sold, and a real content product sits beneath them. CaseMine's own account is that keyword and citation search are free and usable without an account, while AMICUS, CaseIQ, summarisation, argument generation, drafting and contract analysis all sit behind a subscription. That split is the axis in miniature: strip the AI away and what remains is a searchable case law repository across three jurisdictions, which is a product this market has bought for decades and which CaseMine has been building since 2013. The AI layer is substantial and is what the buyer pays for, with retrieval-augmented generation over that repository, an extractive engine taking whole documents as queries, and intent-based search. Verified 2 September 2026.

The corpus itself is a model output, which is as central as this axis goes. Every one of the 3.6 million judicial opinions in the product was summarised by AI, and that summarisation is not a feature layered over a database, it is how the database was built. On top of it sit DescrybeLM as a purpose built legal reasoning engine, Cytator as an AI driven citator producing issue level treatment analysis, and an engine layer exposing retrieval and verification tools to external assistants. Remove the models and there is no corpus, no citator and no product, only public domain opinions the company never had rights to sell. Third on this index at A, after Reveal and Jhana.ai.

Grounding is real, documented and described mechanically rather than asserted. CaseMine states that AMICUS runs on a retrieval-augmented generation framework that actively pulls relevant case laws, statutes and legal principles from its database before generating any response, and it draws the contrast with general tools explicitly, saying this ensures the assistant does not just sound convincing. Responses are stated to be linked to source authorities, with the home page putting it as cited authority out and source stays visible. CaseIQ is separately described as extractive rather than generative, which is a meaningful distinction to publish. What is absent is measurement. No accuracy figure, test set, evaluation or benchmark result appears on any surface read. One thing exists nearby and is recorded rather than credited: trade press reporting the launch of an advanced AMICUS model in March 2026 states that the company maintains an internal benchmarking framework evaluating AI systems across legal research, document analysis, drafting, issue spotting and complex legal reasoning. That is a vendor claim relayed by a third party, no results are published, and it was not located on any first-party page. Verified 2 September 2026.

The best structured grounding disclosure in this category and still short of a published measurement. DescrybeLM answers carry a table of authorities with citations linked back to the primary law, the reasoning is stated to be visible at every step, and the product includes quote verification and citation resolution as first class tools rather than as claims. The vendor also discloses against its own interest: the terms of service state plainly that Descrybe does not review or validate outputs to confirm they are accurate, truthful, relevant, reliable or not misleading, and earlier product material warned users not to assume summary accuracy without checking against the original opinion. A vendor telling a reader that its output is unverified is doing the opposite of overclaiming, and it is rare enough on this index to name. Held at B on two gaps. No measured accuracy figure of any kind is published by the vendor: no summarisation fidelity rate, no citation resolution accuracy, no hallucination rate, no test set. And a benchmark claim is circulating in third party material that DescrybeLM outperforms named frontier models on bar exam benchmarks, which was not located in vendor material with any methodology, evaluator or result table attached, and is therefore not credited here. Checked the home page, the FAQ, the user guide, the terms of service and the product pages on 29 Aug 2026.

Verification is offered as the control and nothing describes an oversight model around it. The published position is that answers carry linked source authority and that the source stays visible, so a user can open what the system relied on, and the vendor frames this as what makes responses verifiable and dependable. That is a real review affordance for a research product. Beyond it, nothing. No published material states what any tool does without a human, where a review point sits, what happens when the system is uncertain, or what a user is expected to check before relying on a draft. This matters more than usual here because AMICUS does not stop at research: CaseMine states it drafts contracts, memos, motions, petitions and transactional documents, and can include judgment-based citations in those drafts. A product that writes filings with citations in them and publishes no oversight model is asserting verifiability without describing where verification happens. Checked the FAQ, the CaseIQ page, the pricing page and the privacy policy on 2 September 2026. Verified 2 September 2026.

Oversight is described as product behaviour rather than as policy, which is the useful form. DescrybeLM is documented to clarify jurisdiction, posture and key facts with the user before proceeding, which puts a human checkpoint at the start of the task rather than only at the end, and it is documented to tell the user when it needs more information and to flag uncertainty rather than filling gaps with guesses. Reasoning is visible at each step and work product is exported deliberately to Word rather than acted on. In the engine configuration the user can see which Descrybe tools an external assistant called and read the data each returned, which is oversight of a tool chain and not merely of an answer. Held at B because none of it is quantified or bounded: no statement of when the system proceeds without clarification, no confidence threshold, and no description of what the uncertainty flag actually triggers.

Basis rewritten 2 September 2026. The original note rested this grade partly on the absence of a customer logo strip, inferred from six first-party pages plus a search extract of a home page that could not be rendered. That reasoning is withdrawn: an active bot block defeats retrieval for any agent, the operator confirmed the same result independently, and a retrieval failure is not evidence of absence. The grade stands on affirmative evidence instead. CaseMine's own competitor comparison page carries an unquantified, unattributed volume claim, that the platform is used by thousands of US lawyers every day, with no named customer, no figure, no date and no case study attached to it anywhere. An unattributed volume claim standing in place of a reference is the finding, and it meets this band's words directly rather than by inference. Nothing located across the FAQ, pricing page, CaseIQ page, privacy policy, about page, bespoke AI solutions page or comparison page names a single customer, quotes a single identified user, or attaches a measured outcome to a named firm. The one institutional relationship the company does publicise is a legal technology certificate programme with an Indian law school, which is a teaching partnership rather than a deployment. Verified 2 September 2026.

Adoption claims are institutional and specific, and none of them was located in vendor material. Third party research material states that Descrybe was added to the NSLT curriculum replacing Casetext, and selected as an approved NELLCO e-resource reaching nearly 150 law libraries. If published by the vendor with detail those would be strong evidence, because a law library consortium and a curriculum adoption are procurement decisions by institutions that evaluate research tools for a living. As located they are third party assertions. Independently verifiable and dated: sustained coverage by a named legal technology journalist across four separate product stages from 2024 to 2026, an ABA Woman of Legal Tech recognition for cofounder Kara Peterson in 2024, and an Anthem Award for Responsible Technology. Held at C because no vendor published customer, deployment, usage figure or outcome measure was located, and because the corpus milestones the vendor does publish describe its own build progress rather than any customer result. Source basis Third Party Estimated on that footing.

One pathway is addressed with real precision and the rest is general. The precise part is CaseIQ: the privacy policy states that nothing uploaded there is stored on CaseMine's server and that any uploaded content is immediately removed after processing, protected by 256-bit end-to-end encryption. That is the strongest single confidentiality commitment on the record. Against it sits the treatment of everything else, in the same document: files, documents and data uploaded to a CaseMine account may be retained on its servers even after deletion or termination of the account, for backup and legal compliance purposes. Those two positions are both published and they point in opposite directions depending on which door a lawyer's document came through. Three limbs are unaddressed entirely. Nothing describes segregation between customers, users or matters. Nothing addresses privilege or work product treatment. And nothing states what the underlying model provider may retain, which matters because the assistant is described as powered by GPT. The no-training commitment is a policy statement and is qualified, as set out on the training signal. Verified 2 September 2026.

Binding published terms exist and nothing in them addresses privilege. The terms of service are framed explicitly as the legal equivalent of a signed written contract, incorporate the privacy policy, and govern the platform rather than merely the website, which puts this vendor ahead of several records in this pull where the only published document was a marketing site disclaimer. Against that: no treatment of legal professional privilege or work product was located, and nothing addresses the confidentiality of the briefs, pleadings and documents that DescrybeLM and Descrybe Review accept as inputs, which is a change in posture from a product that once only served public domain opinions and now ingests a user's own filings. Research limitation stated rather than hidden: a privacy policy exists and is referenced from the FAQ, and its text was not read on 29 Aug 2026. This grade and the AI Safety grade are both correction candidates on that basis.

Row completed 2 September 2026 under R26. A real advice-line position is published, and the vendor's own marketing contradicts it. The AMICUS product page states that information provided by AMICUS should not be considered legal advice; that AMICUS may occasionally include inaccuracies relating to cases, titles or citation codes; that users are advised to consult qualified legal professionals and refer to original records; and that CaseMine is not liable for damages resulting from decisions or actions taken on the information AMICUS provides. That is more than a footer. Naming citation codes as a specific failure mode is unusually candid for a research product, and directing users back to original records is a verification instruction rather than a disclaimer. Against it, the home page markets AMICUS as engineered to always provide dependable advice backed by credible sources and as a trustworthy legal assistant delivering accurate, precise and strategic responses. Both halves belong on the record: the same product is sold as dependable advice and disclaimed as not legal advice that may be inaccurate. Under the rule for a vendor's own marketing statements conflicting with each other, the conflict is graded by asking which limb it shows to be missing, and what it shows is that the position is stated in one place and undercut in another rather than held consistently. Two limbs fail outright: no jurisdiction limits are stated for advice purposes, which matters for a platform sold into India, the United States and the United Kingdom at once, and no bar, Bar Council or law society guidance is engaged anywhere. Provenance limit, recorded rather than hidden: casemine.com serves an active bot block to automated retrieval on every path, for the operator as well as for this reader, so this clause text was recovered through the search index of the first-party URL rather than from a rendered page. No evidence quote is recorded for that reason.

Not addressed, and this is the record on the index where it most obviously should be. The vendor states directly that the product brings legal reasoning to anyone who needs it, lawyers and non lawyers alike, and the founding mission is access to justice for members of the public without formal legal training. A guided reasoning engine that takes a member of the public's question and returns a citation backed analysis is the unauthorised practice question in its clearest form, and no position on it was located: no statement that output is not legal advice, no guidance on when to consult a lawyer, and no engagement with any bar rule. The closest thing is the terms disclaiming that outputs are reviewed or validated, which allocates risk without addressing the professional question. Checked the home page, the FAQ, the user guide, the terms of service and the product pages on 29 Aug 2026.

Nothing published addresses governance over model behaviour. No responsible AI page exists, no governance framework, no individual or function named as accountable, no account of what is tested before a model change ships, and nothing whatsoever about uneven output across jurisdictions, court levels, party types or populations. The certifications the vendor does hold are adjacent subjects this axis treats separately: ISO 9001:2015 is quality management and ISO/IEC 27001:2022 is information security. The one governance-adjacent fact located is reported by trade press rather than published by the vendor, namely an internal benchmarking framework used to evaluate AI systems before deployment; no first-party page describes it and no results are published, so it is recorded on the citation accuracy note rather than credited here. Checked the FAQ, pricing page, CaseIQ page, privacy policy, about page and bespoke AI solutions page on 2 September 2026. Verified 2 September 2026.

Nothing published. No AI policy, no evaluation methodology, no bias or fairness assessment, no accuracy monitoring, no drift statement, no model card, no named governance body and no external standard. The gap has a specific shape here that is worth recording: the entire corpus is machine generated summaries of judicial opinions, so any systematic bias in summarisation propagates into every search result and every citator treatment downstream, and it would be invisible to a user reading the summary rather than the opinion. The vendor rebuilt all its summaries once already, on newer models, which is an implicit acknowledgement that summary quality is model dependent and improvable, and no evaluation of either generation was published. Checked the home page, the FAQ, the user guide, the terms of service and the published product announcements on 29 Aug 2026.

Two of the five elements are published, two are absent and one is generic. Retention is genuinely specific on one pathway, with CaseIQ uploads immediately removed after processing and never stored, and genuinely open-ended on the other, with account content retained on servers even after account deletion or termination for backup and legal compliance. Deletion and data subject rights are set out properly, with access, rectification, erasure, restriction, portability and objection, a thirty day response commitment and appointed EU and UK Article 27 representatives named as Euverify Ltd. Encryption is stated in transit and at rest, with 256-bit AES and SSL or TLS named on the pricing page, and ISO/IEC 27001:2022 certification is claimed. What is missing is what a security reviewer asks for next. No subprocessor list of any kind exists: third parties are given only as categories, being data storage and infrastructure hosting, database and account management, web analytics and payment processing, with not one entity named. And no incident or breach notification practice was located on any surface, which is a notable gap alongside an ISO 27001 certification. Access control is described only as authorised personnel bound by confidentiality. Verified 2 September 2026.

No stewardship position was located for user submitted content. DescrybeLM and Descrybe Review accept briefs, pleadings and documents as inputs, and research sessions are saved and revisited as a product feature, so user content is both ingested and persisted by design. Nothing located states whether that content is used to train or improve any model, how long it is retained, whether a user can delete it, or how it is handled when passed to the third party AI models the terms disclose are in use. Same research limitation as Privilege: the privacy policy exists, is referenced from the FAQ, and was not read on 29 Aug 2026, so this grade is a correction candidate and the absence is recorded against the material actually checked, being the home page, the FAQ, the user guide, the terms of service and the product pages.

Row completed 2 September 2026 under R26, and decided rather than held on the reasoning below. Liability is addressed, once, and only by disclaiming the exposure the product creates. The AMICUS product page states that CaseMine is not liable for damages resulting from decisions or actions taken on the information AMICUS provides, alongside an acknowledgement that output may contain inaccuracies relating to cases, titles or citation codes. That is a published position on who bears the loss when the output is wrong, and the answer is the user, which is why the bottom band no longer fits: something is published. Nothing else is. No liability cap, no indemnity in either direction, no warranty, no service level and no insurance is published on any reachable surface. The Terms of Use referenced by the privacy policy could not be located by either this reader or the operator, so the contractual allocation a signed customer actually receives is unknown, and under the rule that credit follows scope a disclaimer on a product page is not the agreement. What is graded here is the state of the published record, not an inference about the contract. Two things would move this row: the Terms of Use surfacing, or a cap or indemnity appearing anywhere. Provenance limit: casemine.com serves an active bot block on every path, so this text was recovered through the search index of the first-party URL rather than a rendered page, and no evidence quote is recorded for that reason.

An actual published liability position exists, which most of this roster lacks, and it runs almost entirely one way. The terms of service state that the user indemnifies and holds Descrybe harmless from liability, loss, claim and expense including reasonable attorneys' fees related to use of the service or its outputs, and that any liability found on Descrybe's part is limited to the amount the user actually paid in the twelve months before the action giving rise to it. Read against published pricing that caps vendor exposure at $300 to $600 per user. The same document states that Descrybe does not review or validate outputs for accuracy. Graded C rather than D because a specific, quantified and readable cap is genuine disclosure that a buyer can price, and it is more than the vendors on this index who publish nothing. No warranty, accuracy undertaking, service level or correction mechanism was located.

No integration into the systems legal work already lives in was located, and none is claimed. Checked the FAQ, the pricing page and its full feature list, the CaseIQ page, the privacy policy, the about page and the bespoke AI solutions page on 2 September 2026. There is no integrations page, no named connector, no API or developer documentation, and no mention of document management, practice management, Microsoft Word, Outlook or e-filing anywhere. The vendor addresses the question directly and answers it the other way: asked whether anything needs to be installed, it states that CaseMine is fully cloud-based and accessed through a browser on desktop and mobile with no downloads or setup, which is a statement that the product is self-contained rather than connected. The nearest thing to integration is document upload, which is a file transfer rather than a connection, and the bespoke solutions page offers to build tooling around a firm's documents and workflows on a consulting basis rather than as a published integration. Verified 2 September 2026.

Integration is the product rather than an afterthought, and it is documented at tool level. Descrybe Legal Engine exposes citation lookup, quote verification, treatment review and source retrieval to external assistants over an MCP interface, with named supported hosts including Claude, ChatGPT and Perplexity, and an Open Connector for building further tools. A user guide describes the call pattern: the assistant invokes one or more Descrybe tools, Descrybe performs the search, lookup, verification or retrieval, and returns focused results, with the user able to see which tools were called and read what each returned. Work product exports to Word. Held at B because the integrations are to AI assistants rather than to legal practice systems: no document management system, no practice management platform, and nothing connecting to a firm's existing matter estate. Notable as the first record in this pull whose integration story is an interoperability protocol rather than a connector list.

Cloud delivery is stated plainly and neither limb this axis asks for follows. The vendor confirms the platform is fully cloud-based and browser-accessed with no installation, so the delivery model is not in doubt. But no region is published anywhere: the privacy policy says data is held on secure servers without saying where those servers are, and the only geography stated is corporate rather than technical, namely that operations are based in India with services provided across India, the United States and the United Kingdom. No residency option is offered, no processing location is distinguished from storage location, and nothing addresses where data of a UK or US customer physically sits. Nor is any tenancy model stated: nothing describes the platform as multi-tenant, single-tenant or otherwise isolated. The appointment of Euverify Ltd as Article 27 representative for the EU and UK indicates cross-border processing is contemplated without describing it. Verified 2 September 2026.

Nothing located. No hosting provider, no region, no data residency commitment, no single tenant or private option, and no statement of where user submitted briefs and pleadings are processed or stored. The engine architecture makes this more pointed than usual rather than less: user queries and returned legal material move between the Descrybe service and a third party assistant host such as Claude, ChatGPT or Perplexity, and nothing published describes what crosses that boundary, where either side sits, or which party holds what. Checked the home page, the FAQ, the user guide, the terms of service and the product pages on 29 Aug 2026.

Certification is real, stated and named to the version, which is more precise than most records manage: CaseMine states it is certified under ISO 9001:2015 for quality management and ISO/IEC 27001:2022 for information security. Encryption specifics are published alongside, with 256-bit AES and SSL or TLS listed as a feature of every subscription tier and end-to-end encryption stated for CaseIQ uploads. Everything the top band asks for beyond the standard name is absent. No certification body or auditor is named, no certificate number, issue date or expiry is published, no scope statement is given for either standard, and no route to obtain evidence is offered, whether self-serve, on request or under a non-disclosure agreement. There is no trust centre or portal of any kind, no penetration testing or vulnerability assessment programme is described, and no SOC 2 report is claimed. A buyer is asked to take two certifications on the strength of a sentence. Verified 2 September 2026.

No certification, attestation or security documentation of any kind was located. No SOC 2 of either type, no ISO 27001, no penetration testing statement, no named auditor, and no trust centre, security page or documentation request route. Under the three tier test the artifact is absent rather than gated. Some context without changing the grade: this is a self funded two founder company that priced the product at $25 to $50 per month and states it carries no enterprise sales overhead, and a SOC 2 examination is a material cost against that model. The index grades what a buyer can verify, so the grade stands, and the note records why the absence is unsurprising rather than treating it as concealment. Checked the home page, the FAQ, the user guide, the terms of service and the site footer on 29 Aug 2026.

The architecture is described well and the supply chain behind it is not. On architecture the disclosure is better than most: retrieval-augmented generation is named as the framework, the sequence is described, with relevant case law, statutes and principles pulled from CaseMine's database before any response is generated, and CaseIQ is separately identified as extractive rather than generative. The model layer is named only at the level of a family, with AMICUS described repeatedly as powered by GPT and the premium tier promising the latest and most advanced AI models. That tells a reader roughly whose technology is underneath and nothing more. No provider entity is named, no model or version is identified, nothing states where inference runs or what the provider may retain, and no change notification commitment exists despite the premium tier being sold partly on receiving newer models. The privacy policy lists third parties by category only, so the supply chain cannot be reconstructed from there either. Verified 2 September 2026.

The category is disclosed in the binding document and no provider is named, and a third party claim to the contrary was rejected. The terms of service state that Descrybe uses third party AI models to summarise and make searchable primary law and judicial cases. That is a real supply chain disclosure, made in the contract rather than in marketing, and it establishes that outside models process the corpus. Third party research material asserts the opposite, that DescrybeLM is the company's own model and is not reliant on third party models such as ChatGPT or Claude. Where a directory and a vendor's own terms of service conflict, the terms govern, and the third party claim is not credited. Held at C because no provider, model or version is named, no subprocessor list exists, and nothing states which models touch user submitted briefs as opposed to the public opinion corpus. Compare Onspring at B, the only record on this index naming its provider outright.

Amended 2 September 2026: the figures were recovered and the grade moves from B to A. A buyer can learn what this costs without entering a sales process. Three price points are published against named tiers with full inclusions: 99 US dollars per month for AMICUS AI with 250 credits, CaseIQ, CiteText, 50MB document upload, up to 20 cases and 2GB storage; 149.99 US dollars per month for 500 credits plus 100MB uploads, web search, the latest models, a dedicated account manager and unlimited cases; and an annual rate of 1,499.99 US dollars shown against a struck-through 1,799.99. The unit being charged is explicit and unusually granular, running to AI credits per month, upload size, storage and a cap on matters. A genuine free tier sits beneath all of it, with keyword, citation and statute search plus full judgment texts usable without any account, and a separate free student registration. The vendor states no commitment and no fees. Two limitations are recorded as limits on this reading rather than as findings against the vendor, because casemine.com serves an active bot block to automated retrieval on every path and these figures were recovered through the search index rather than a rendered page. The price of the entry Essential tier was not among the figures recovered, so whether it is free or paid is unestablished. And two readings of the same page disagree on the top tier's storage, one giving 5GB and the other unlimited. Neither is a gap in what the vendor publishes. A third-party software directory claims paid plans start at 149.99 and that a free-forever plan exists; first-party evidence beats it, the 99 dollar rate stands, and the conflict is noted rather than relied on.

The first A on this axis in the pull, and it is earned on the full set. Exact prices are published for every tier: $25 per month for Descrybe Legal Engine and $50 per month for the full Descrybe Platform with the Engine included. The unit of charge is a monthly subscription per user, stated plainly. Feature gating is explicitly ruled out, with the vendor stating that every feature is included at either tier, which removes the usual variable where a headline price buys an unusable subset. The vendor goes further and explains the pricing basis, stating that it is self funded, without outside investor timelines or enterprise sales overhead, and prices for the work rather than the market. A buyer can determine total cost of ownership from the public site without contacting anyone, which is the top of this axis. Every other record in this pull to date requires a demo request or a login.

Coverage is documented jurisdiction by jurisdiction and court by court, which is the most precise treatment of this axis in the lane. India is stated as the Supreme Court, all High Courts, tribunals and commissions. The United States is stated as all fifty states and federal jurisdictions, naming the Supreme Court, federal circuit courts, district courts and the specialised bankruptcy and immigration courts, and distinguishing complete appellate and supreme court coverage from trial court rulings in select states, which is an honest limitation rather than a blanket claim. The United Kingdom is stated as the Supreme Court, Court of Appeal, High Court and major tribunals. The buyer is named per market too, running to advocates, attorneys, solo practitioners, solicitors, barristers, law firms, legal researchers and students. Two limbs are missing. Nothing addresses in-house legal departments or government use, both absent from every buyer list. And practice areas are not enumerated at all, which is defensible for a case law platform that is practice-agnostic by design but leaves a buyer without a statement of where the product stops. Verified 2 September 2026.

Coverage is stated concretely and its boundaries are honest. More than 3.6 million judicial opinions covering state supreme and appellate courts across all fifty states and the District of Columbia, extended to federal district and appellate opinions, with statutes and regulations also addressed by the current platform. Corpus growth was published as it happened, from seven states and 1.2 million opinions to national coverage, which is a checkable record rather than a static claim. Bilingual English and Spanish search and summaries are a genuine coverage dimension almost nothing else on this index offers, and plain language versions extend reach to non specialists. Held at B rather than A because the corpus is not enumerated at the level a researcher checks: no per court or per jurisdiction breakdown, no historical date range, and no update lag or refresh frequency, so a practitioner cannot confirm whether last quarter's appellate decision in their state is present.

The commitment sits in the security answer of the public FAQ and the qualifier in it is the finding. The promise is that data is never used to train external models, which addresses the third-party layer and leaves CaseMine's own models unaddressed. Given the assistant is described throughout as powered by GPT, the natural reading is that customer content does not flow into the base model provider's training, which is a real and useful commitment; what no published material states is whether customer content informs CaseMine's own retrieval, embedding or fine-tuning layers. No agreement could be tested against it. The privacy policy is silent on model training in either direction, and the Terms of Use it references could not be located or retrieved on 2 September 2026. Two adjacent statements are recorded because a reader will meet them and neither concerns customer content: AMICUS is described as trained entirely on authoritative legal sources and exhaustively trained on comprehensive legal archives, which describes the corpus rather than user data. Under the naming test the value rests on the published policy statement, qualified as it is.

Silent on the question that now matters. The quoted clause from the terms of service establishes that outside models process material, but it is scoped to summarising and making searchable primary law and judicial cases, which is public domain content. It says nothing about the briefs, pleadings and documents that DescrybeLM and Descrybe Review accept from users, and no statement was located anywhere addressing whether user submitted content or research queries are used to train or improve any model, in either direction. Recorded as silent, not as a negative commitment. Research limitation: a privacy policy exists and is referenced from the FAQ, and its text was not read on 29 Aug 2026, so this value is a correction candidate. Checked the terms of service, the home page, the FAQ, the user guide and the product pages.

Two pathways are published and they point in opposite directions, which is why the overall position is acknowledged rather than settled. The quoted commitment continues that any content uploaded in CaseIQ is immediately removed after processing, protected by 256-bit end-to-end encryption, and that is as tight a retention position as anything in this pull. The other pathway is open-ended: the same privacy policy states that files, documents and data uploaded to a CaseMine account may be retained on secure servers even after deletion or termination of the account, for backup and legal compliance purposes, with no period attached. Nothing states how long prompts, chats or generated drafts persist in the ordinary course, and no retention setting is offered to the customer. Deletion is available on request with a thirty day response commitment, and data subject erasure rights are set out. A buyer needs both halves: the same document a lawyer drops into CaseIQ is discarded at once, while the same document saved into their account may outlive the account itself.

Not addressed. Retention is built into the product by design, since research sessions are saved and revisited as a documented feature, so user queries and generated output persist. No period is stated, nothing indicates whether a user can delete a saved session or purge history, and nothing distinguishes retention of a research query from retention of an uploaded brief. Same privacy policy limitation as above applies and this value is a correction candidate on the same footing. Checked the terms of service, the FAQ, the user guide and the product pages on 29 Aug 2026.

Checked the FAQ, the pricing page and its full feature list, the CaseIQ page, the privacy policy, the about page and the bespoke AI solutions page on 2 September 2026. Nothing describes segregation between customers, between users inside a firm, or between matters. No tenancy or isolation statement appears anywhere. What exists is adjacent and concerns the vendor's own staff rather than the customer's structure: access to personal information is stated to be limited to authorised CaseMine personnel who require it and who are bound by confidentiality obligations, and a separate clause reserves access to account contents by employees or service providers when investigating suspected illegal activity or terms violations. The product does have matter-shaped features, with case management capped by tier at five, twenty or unlimited cases and personalised folders for storing documents, but nothing describes those as a security boundary.

Not addressed. No permission model, access control or segregation description was located. The product is sold as a per user subscription rather than as a firm deployment, so there is no documented firm workspace in which walls would operate, and nothing describes what happens where several users at one firm subscribe. No document management system integration exists to inherit permissions from. Checked the home page, the FAQ, the user guide, the terms of service and the product pages on 29 Aug 2026.

The privacy policy addresses compelled disclosure squarely and never reaches notice. It states that personal information may be disclosed when required by law or in response to the legal processes quoted, to comply with legal obligations in the jurisdictions where CaseMine operates, and to law enforcement or government authorities where the company determines disclosure is necessary to protect the safety, rights or property of users, employees, the public or CaseMine itself. That last limb is a unilateral determination rather than a legal compulsion. Nothing anywhere commits to telling the customer a demand has arrived, and nothing reserves discretion over notice either, so the question is not reached rather than answered adversely. Checked the privacy policy in full, the FAQ, the CaseIQ page and the pricing page on 2 September 2026. The Terms of Use referenced by the privacy policy could not be located or retrieved and might address it.

Not addressed. No government or law enforcement request clause, no commitment to notify a user before producing their data, and no transparency report were located. The terms of service address termination of access, permitted use and liability, and do not reach third party requests. Checked the terms of service, the FAQ and the site footer on 29 Aug 2026. Privacy policy unread, same correction candidate footing as the rows above.

The corpus is identified court by court rather than by publisher, which for primary case law is identification of the sources themselves. India is given as the Supreme Court, all High Courts, tribunals and commissions. The United States is given as the Supreme Court, federal circuit courts, district courts, bankruptcy and immigration courts, and all fifty states, with a stated distinction between complete appellate and supreme court coverage and trial court rulings in select states only. The United Kingdom is given as the Supreme Court, Court of Appeal, High Court and major tribunals. Update cadence is described rather than claimed, as continuous integration of newly published rulings through an automated data pipeline, stated separately for each jurisdiction. What is absent is the rights position. No licence, contract or permission basis is stated for any jurisdiction's material, no publisher or data supplier is named, and nothing addresses the secondary material the platform also carries in commentaries and columns. The company describes a long-term focus on legal data infrastructure without describing how the data is obtained.

Sources named at dataset level with no licence basis stated, and this is among the better provenance records in the category. The Harvard Caselaw Access Project is named as a source, which is a specific and checkable dataset rather than a vague reference to public records, and the corpus is described as state supreme and appellate opinions across all fifty states and the District of Columbia extended to federal district and appellate opinions, with statutes and regulations in the current platform. Coverage growth was published in stages as it was built, which gives a reader a dated trail. Not stated: the licence or public domain basis on which any source is used, a per court enumeration, a historical date range, or an update lag. The corpus consists of AI generated summaries of the underlying opinions rather than the opinions alone, so provenance here has a second limb the vendor does not address, being what the summarisation step did to the source.

CaseMine publishes its own citator rather than licensing one, and this is the first record in this pull to reach that value. The FAQ describes a built-in citation analysis feature called the CaseMine Citator which tracks how a case has been treated in subsequent decisions, flags negative treatment and overruling as quoted, and reports where a judgment has been cited and how many times, with the stated purpose of letting a user assess whether a precedent is still good law. It is not an add-on: the pricing page lists checking for negatively treated and overruled judgments with an advanced citator as a feature of the entry tier, so every paying user has it, and an importance matrix and citation search sit alongside. What is not published is anything about how the treatment signal is produced, whether editorially or automatically, what its categories are beyond negative treatment and overruling, or how completely it covers each of the three jurisdictions. A vendor-built signal with an undescribed method is what separates this value from the licensed tier above it.

Own treatment signal, and the most developed one located in this pull. Cytator launched June 2025 as part of the Legal Research Toolkit, presenting search results with treatment flags in the familiar convention of positive, negative, cautionary and neutral, and offering issue level analysis rather than a single case level verdict. Published detail includes forward treatment showing binding citing cases and non binding cited cases, and a backward citator showing how the opinion treated each authority it relied on. Treatment review is also exposed as a tool to external assistants through Descrybe Legal Engine. It is the vendor's own AI derived citator rather than a licensed commercial citator such as Shepard's or KeyCite, and no accuracy, coverage or agreement measure against an established citator is published, which is the gap a practitioner would want closed before relying on a negative flag.

Checked the FAQ, the CaseIQ page, the pricing page and the privacy policy on 2 September 2026. Nothing describes what the platform does when it cannot ground an answer, and no confidence, relevance or grounding indicator is described as shown to the user. The architectural answer offered instead is that retrieval happens before generation, so that responses are grounded in real legal sources rather than sounding convincing, and that citations link back to source authority the user can open. That is a design claim about how hallucination is reduced rather than an account of what the system does at the point where retrieval returns nothing useful. Given the product drafts motions and petitions with judgment-based citations included, what happens when authority is thin is a live question the published material does not reach.

Documented, and one of very few records in the pull to reach this value. The vendor states that DescrybeLM tells the user when it needs more information and flags uncertainty rather than filling gaps with guesses, and separately that the engine clarifies jurisdiction, posture and key facts before proceeding rather than answering an underspecified question. That is an explicit no answer path described as product behaviour. The terms of service reinforce it from the other direction by stating that outputs are not reviewed or validated for accuracy. Recorded as documented rather than documented and demonstrable because no published evaluation, transcript or example shows the behaviour operating, and no threshold or trigger for the uncertainty flag is described.

Searched the AI Hallucination Cases database maintained by Damien Charlotin, and reporting drawing on it, on 2 September 2026 on the product name CaseMine, the assistant name AMICUS and the corporate name Gauge Data Solutions. No court order, opinion or disciplinary record naming the product was located. This is a statement about the public record rather than a finding about the product. Two things bound it. The database is heavily weighted to United States filings and its Indian coverage is thin, while India is CaseMine's home market and the market where fabricated-citation incidents have recently reached the High Courts and the Supreme Court. And Indian legal commentary published in July 2026 discusses AMICUS alongside other Indian retrieval-grounded tools as reducing rather than eliminating hallucination, which is analysis rather than a court finding and is recorded here only because a reader searching this ground will meet it.

None located, with the instrument named. General web searches combining the vendor and product names with court, opinion, sanction, hallucination and fabricated citation terms returned nothing on 29 Aug 2026. No named docket database or court record tracker was searched. Recorded as a statement about what this search found and not as a clearance. Worth noting for a later reader that the vendor itself discloses in its terms that it does not validate outputs, so a defective citation is a disclosed possibility rather than one the vendor denies, which is a different posture from most of the roster.

Checked the FAQ, the pricing page, the CaseIQ page, the privacy policy, the about page and the bespoke AI solutions page on 2 September 2026. No public material engages with professional or ethics guidance from any of the three markets served. Neither ABA Formal Opinion 512 nor any state bar opinion is named for the United States, no Bar Council of India or Solicitors Regulation Authority material is named, and no equivalent guidance appears for barristers or solicitors in the United Kingdom. The gap is more pointed than usual because the platform is sold into three regulatory systems at once and drafts filings in all of them, and because the vendor also runs a legal technology certificate programme with an Indian law school, so it engages the profession's institutions in a training capacity without engaging their guidance in the product.

Not addressed. No named ethics opinion, no ABA Formal Opinion 512, no state bar guidance and no engagement with professional conduct rules was located, despite the cofounder holding an ABA Woman of Legal Tech recognition and the company positioning itself around access to justice. The absence is most visible where the product is offered to non lawyers alongside lawyers, since that is the configuration bar guidance speaks to most directly. Checked the home page, the FAQ, the user guide, the terms of service and the product pages on 29 Aug 2026.

Checked the FAQ, the pricing page, the CaseIQ page and the privacy policy on 2 September 2026. Nothing addresses how AI-assisted work is recorded, billed or disclosed to a client, and no per matter record of AI-assisted work was located. Unusually for this corpus there is also no quantified savings claim to record: the vendor's language is qualitative throughout, describing speed, clarity and working smarter without publishing a time or cost figure anywhere, so the weakest value is not made out either. The product does count AI credits per user per month and does organise work into cases, so the raw material for a per matter record of AI use exists in the platform, but nothing describes it as reportable or exportable, and credits are presented as a consumption allowance rather than a record.

Not addressed. Nothing published addresses billing for AI assisted time or describes a record a practitioner could disclose to a client showing what was machine generated. The vendor is unusually transparent about what it charges the subscriber, at $25 and $50 per month with all features included, and that is the vendor's own price rather than a position on the practitioner's side of the equation. No time record, audit record or fee guidance was located. Checked the home page, the FAQ and the terms of service on 29 Aug 2026.

Checked the FAQ, the pricing page, the CaseIQ page, the privacy policy in full, the about page and the bespoke AI solutions page on 2 September 2026. No subprocessor list exists in any form. The privacy policy names categories only, listing data storage and infrastructure hosting, database and account management, web analytics and payment processing, without identifying a single entity behind any of them. No model provider is named as a legal entity, the assistant being described only as powered by GPT. No data processing agreement is published, no forwardable client-facing disclosure material was located, and there is no trust centre or review pack. A firm asked by a client which entities may see its matter documents could not answer from anything CaseMine publishes. The two ISO certifications are claimed without any accompanying scope or report, so they do not fill the gap either.

Not addressed. No subprocessor list, no named model provider, no security documentation, no trust centre and no request route were located, so a firm has nothing it could forward to a client. The terms of service do disclose that third party AI models are used, which is the category and not the entity, and is the single element of such a pack that exists. Checked the terms of service, the FAQ, the user guide and the site footer on 29 Aug 2026.

Checked the FAQ, the pricing page and its full feature list, the CaseIQ page and the privacy policy on 2 September 2026. Nothing addresses judicial standing orders, disclosure of AI use, or certification that citations were verified by a person. The platform does produce artifacts a professional could keep, with citations linked to source authority, an unlimited search history at every tier, task management and case folders, but none is described as a record of model use, and nothing captures which model produced which passage or who reviewed it. The gap is worth naming on this record in particular: the product drafts motions and petitions with judgment-based citations embedded, in three jurisdictions where courts have begun asking exactly this question, and the citator it publishes is the closest thing it offers to a verification trail.

Partial record, and stronger on the sources limb than most. DescrybeLM answers carry a table of authorities with verified citations linked back to the primary law, the reasoning is stated to be visible at every step, quote verification and citation resolution are available as discrete tools, research sessions are saved and revisitable, and work product exports to Word, so a practitioner can produce what was retrieved and relied on. In the engine configuration the user can additionally see which tools an assistant called and read what each returned, which is a tool level trail few products expose. What is missing is the other two limbs a standing order asks for: nothing records which model produced a given output, and no human verification record is captured or exportable. The vendor's disclosure that it does not validate outputs makes the human verification gap more consequential, since the checking is entirely the practitioner's and nothing in the product evidences that it happened.