Descrybe is a US primary law research platform built on AI summaries of judicial opinions, now at descrybe.com (descrybe.ai redirects). It was founded in July 2023 by Kara Peterson and Richard DiBona near Boston, is self funded, and began as a free service aimed at opening legal information to smaller firms, journalists and the public. Its coverage page, with data as of 18 July 2026, lists 10,969,112 opinions across 492 federal, state, bankruptcy, specialized, agency and tribal courts and tribunals, plus statutes, regulations, constitutions, session laws and court rules, each court with its own count and date range.
Summaries and search work in English and Spanish, with plain language versions. The Descrybe Platform includes DescrybeLM, launched March 2026, a guided legal reasoning engine that clarifies jurisdiction, posture and facts and returns citation backed answers with a table of authorities, along with brief checking, search, citation and quote verification, and the Cytator citator. Descrybe Legal Engine offers citation lookup, quote verification, treatment review and source retrieval to assistants such as Claude, ChatGPT and Perplexity through MCP.
Pricing is $25 a month for the Engine and $50 for the full Platform. A security page names OpenAI's API as the main model provider and DigitalOcean's New York region for storage.
Side by side on the same 15 capability axes and 12 legal signals: where the two part company, and what neither one publishes.
Vendors in the same categories as Descrybe whose graded profiles look most alike, with a line on what each product does.
Similar is not the same, so check that each one does the job you need. No vendor pays for placement.
Ordered by how closely the graded profiles match across the 15 capability axes, inside the same categories. See how each axis is graded in the methodology.
All 15 axes, graded from public sources on the date shown. Hover a grade to see what the letter means on that axis.
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.
The content itself is model output, which is as central as AI can be. The product was built by AI summarizing judicial opinions, now 10,969,112 opinions across 492 courts and tribunals per the July 2026 coverage page, and that summarization is how the database was built, not a feature on top of it. On top sit DescrybeLM, a purpose built legal reasoning engine; Cytator, an AI citator giving treatment analysis by issue; and an engine layer offering retrieval and verification tools to outside assistants.
Without the models there would be no summaries, no citator and no product, only public domain opinions the company never had rights to sell.
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.
Grounding is well structured, without a published measurement. DescrybeLM answers carry a table of authorities linked to primary law, the reasoning is shown at each step, and quote verification and citation resolution are built in as tools. The vendor also discloses against its own interest: the terms say Descrybe does not review or validate outputs for accuracy, truthfulness, relevance or reliability, or to confirm they are not misleading, and earlier material warned users to check summaries against the original opinion.
No accuracy figure of any kind is published: no summary fidelity rate, citation resolution accuracy, hallucination rate or test set. A claim in third party material that DescrybeLM beats named frontier models on bar exam benchmarks was not found in vendor material with any method or results, and is not relied on.
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.
Oversight is described as product behavior 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. 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.
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.
Institutional adoption claims are specific, but none was found in vendor material. Third party research says Descrybe was added to the NSLT curriculum in place of Casetext and approved as a NELLCO electronic resource reaching nearly 150 law libraries. Those would be strong evidence if the vendor published them, since library consortiums and curricula evaluate research tools carefully, but as found they are third party claims.
Dated and checkable: coverage by a named legal technology journalist across four product stages from 2024 to 2026, an ABA Woman of Legal Tech award for cofounder Kara Peterson in 2024, and an Anthem Award for Responsible Technology. No customer, deployment, usage figure or outcome measure published by the vendor was found, and the coverage milestones describe the vendor's own build, not customer results. The source basis is recorded as third party estimated.
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.
Published commitments cover what firms ask first, but not privilege itself. The security page, updated 30 July 2026 and read 11 September 2026, says customer prompts, files and history are not used to train Descrybe owned models; OpenAI's API is the main model provider, opted out of training, with abuse logs kept up to 30 days; uploaded briefs are not kept once text is extracted; users can delete research history at any time; and the product uses individual accounts with no shared or matter workspaces, so separation is per user and matter walls do not exist.
The binding terms of service, described as the legal equivalent of a signed contract, say nothing on these points, so the commitments are policy, not contract. Privilege and work product are not addressed anywhere, including the privacy policy (updated 29 September 2025) and the security page, though the product now takes in users' briefs and pleadings. Where OpenAI processes requests is not stated.
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.
Not addressed, on a product where it most obviously should be. The vendor says the product brings legal reasoning to anyone who needs it, lawyers and nonlawyers alike, and its founding mission is access to justice for people without legal training. A guided reasoning engine that takes a member of the public's question and returns a cited analysis raises the unauthorized practice question directly, and no position was found: 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 is the terms disclaiming review or validation of outputs, which allocates risk without addressing the professional question.
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.
Nothing is published on governance: no AI policy, evaluation method, bias or fairness assessment, accuracy monitoring, drift statement, model card, governance body or external standard. The gap matters here because every summary is machine generated, so any systematic bias in summarization flows into every search result and citator treatment, unseen by a user who reads the summary rather than the opinion. The vendor has already rebuilt all its summaries once on newer models, showing that quality depends on the model, and published no evaluation of either version.
On 11 September 2026 the privacy policy and the security page, updated 30 July 2026, were also read; the security page says models may be updated for quality, safety, performance or reliability without notice unless the change is material, which concerns the supply chain, not governance.
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.
Most of the ground is covered, specifically enough to hold the vendor to. The security page, updated 30 July 2026 and read 11 September 2026, says what is kept (prompts, research history, outputs, metadata and logs) and what is not (uploaded files are discarded once text is extracted). Users can delete history at any time and accounts on request; no zero retention mode is offered; production access is limited to authorized technical staff on a need to know basis; and data is encrypted in transit and at rest on DigitalOcean managed services in the New York region.
It names subprocessors with links to their attestations: OpenAI's API, DigitalOcean, Clerk and Stripe, with Google Analytics in the privacy policy. OpenAI API training is opted out. Two gaps remain: no incident or breach practice is published, and no retention period is given for logs and metadata outside the user's deletion control.
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.
A published liability position exists and runs almost entirely one way. The terms of service make the user indemnify and hold Descrybe harmless for liability, loss, claims and expenses, including reasonable attorneys' fees, related to use of the service or its outputs, and limit any Descrybe liability to what the user paid in the twelve months before the claim. At published prices, that caps vendor exposure at $300 to $600 per user.
The terms also say Descrybe does not review or validate outputs for accuracy. A specific, readable cap is real disclosure a buyer can price. No warranty, accuracy undertaking, service level or correction route was found.
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.
Integration is the product, documented at tool level. Descrybe Legal Engine offers citation lookup, quote verification, treatment review and source retrieval to outside assistants over MCP, naming Claude, ChatGPT and Perplexity as supported hosts, plus an Open Connector for building more tools. A user guide describes the flow: the assistant calls one or more Descrybe tools, Descrybe runs the search, lookup, verification or retrieval and returns focused results, and the user can see which tools were called and what each returned.
Work exports to Word. The integrations are to AI assistants, not legal practice systems: there is no connection to document management, practice management or a firm's matters.
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.
Tenancy and storage region are stated; processing location is half stated. The security page, updated 30 July 2026 and read 11 September 2026, says customer data is stored on DigitalOcean servers in the New York City region, the product runs on provider managed multitenant infrastructure, it is built around individual user accounts, and no single tenant, private or shared workspace option exists. Research requests are processed by OpenAI's API, and nothing says in which region.
Queries routed through an outside assistant such as Claude, ChatGPT or Perplexity add a leg whose location is also not described. No residency options are offered.
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.
The vendor states plainly that it holds no attestation of its own. The security page, updated 30 July 2026 and read 11 September 2026, says Descrybe does not hold its own SOC 2 or ISO 27001 certification and has begun work toward SOC 2. It relies on providers' programs, linking to trust portals for DigitalOcean (SOC 2 Type II, SOC 3), OpenAI (SOC 2 Type 2, ISO 27001, 27017, 27018, 27701), Stripe (PCI Level 1, SOC 2 Type 2) and Clerk (SOC 2 Type 2).
Data is encrypted in transit over TLS and at rest on DigitalOcean, and production access is need to know. A provider's report covers its infrastructure, not the application layer where Descrybe's code, access controls and data handling sit, which is what a firm's security questionnaire asks about. Descrybe is a self funded two founder company charging $25 to $50 a month, and a SOC 2 examination is a material cost against that model. The security page sits in the site footer.
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.
The provider is named, and change notification is the gap. The security page, updated 30 July 2026 and read 11 September 2026, says Descrybe uses OpenAI API models for summaries and research outputs, that OpenAI's API is its main LLM provider, that the exact model may vary by feature and change over time for quality, safety, performance or reliability, and that users are not notified of each update unless it materially affects terms, privacy or data processing.
This settles earlier conflicting signals: the terms disclose third party models without naming them, and a third party directory had called DescrybeLM the company's own model. The page also says OpenAI training is opted out and abuse logs are kept up to 30 days, linking to OpenAI's data controls. No specific model or version is named, there is no commitment to notify of model changes, and where OpenAI processes requests is not stated.
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.
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 is a monthly subscription per user. The vendor says every feature is included at either tier, so the headline price does not buy a limited subset. It also explains its pricing: it is self funded, without outside investor timelines or enterprise sales costs, and prices for the work rather than the market. A buyer can work out total cost from the public site without contacting anyone.
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.
Coverage is listed court by court, with limits stated. The coverage page (descrybe.com/coverage, data as of 18 July 2026) lists 492 courts and tribunals and 10,969,112 opinions: federal (137 courts, 3,221,272 opinions), state (201 courts, 7,324,869), bankruptcy (97 courts, 80,941), specialized and agency tribunals (47 tribunals, 340,850) and tribal courts, plus statutes, regulations, constitutions, session laws and court rules.
Each court shows its opinion count and first and latest dates, so a researcher can check whether a recent decision should be present. Most active supreme and appellate courts run to the first days of July 2026, and gaps are shown, such as Connecticut Superior Court ending in December 2014 and the Alabama Court of Appeals in 1999. No refresh schedule is stated beyond the as of date, and a few rows, including the New Hampshire district court and the Minnesota and Arizona appellate courts, show latest dates in the future, which look like source metadata. The vendor says it serves lawyers and nonlawyers alike.
The public pages on file for Descrybe, with the recorded signals each one supports and the date it was last read. Open any of them and check the reading against the record.
Client Data in Training, Prompt and Output Retention, Ethical Walls and Matter Segregation and 1 more
Read Sep 11, 2026
Primary Law Corpus Provenance
Read Sep 11, 2026
Exact prices published for every tier: $25 per month for Descrybe Legal Engine and $50 per month for the full Descrybe Platform with the Engine included, as a monthly subscription per user. The vendor states that every feature is included at either tier, ruling out feature gating, and explains its pricing basis as a self funded company without outside investor timelines or enterprise sales overhead. Recorded from the Commercial Transparency read; figures as published on the verification date of that row.
Implementation: Not stated. No onboarding or setup charge is published; the product is self serve at both tiers.
Note: A buyer can work out total cost from the public site without contacting anyone.
Descrybe released its Legal Engine plugin for ChatGPT Enterprise. The company says it gives ChatGPT users structured, verifiable US primary law and legal intelligence built for AI.
Descrybe published a security page stating its data handling position: customer prompts, files and conversational history are not used to train Descrybe owned models; the primary model provider is OpenAI's API, with Descrybe opted out of training and OpenAI abuse monitoring logs retained up to 30 days; uploaded files are discarded once text is extracted; research history can be deleted by the user at any time, with no zero retention mode offered; data is stored on DigitalOcean in the New York region on multi tenant infrastructure, encrypted in transit and at rest; the product runs on individual user accounts with no shared or matter based workspaces; and Descrybe holds no SOC 2 or ISO 27001 certification of its own, with SOC 2 work begun. Subprocessors are listed with links to each provider's attestation.
Descrybe published a coverage page enumerating its primary law corpus court by court: 492 courts and tribunals and 10,969,112 opinions as of 18 July 2026, across federal, state, bankruptcy, specialized and agency, and tribal courts, plus statutes, regulations, constitutions, session laws and court rules, with an opinion count and a first to most recent date range for every court. The vendor's earlier public figure was 3.6 million opinions across state supreme and appellate courts.
A signal records what public sources say on the date shown. It is not a grade and it is not a recommendation. Where a signal reads Not addressed, it means the index did not locate the material in public sources on that date, which is a statement about disclosure rather than about the product.
Can material a lawyer puts into this product be used to train a model?
A public policy or trust page states no training on customer content, with no matching term located in the published agreement.
Stated in policy, in the form that matters. The security page, updated 30 July 2026 and read 11 September 2026, says Descrybe does not use customer prompts, files or conversational history to train Descrybe owned models, that its main model provider is OpenAI's API, that API inputs are not used for OpenAI training by default, and that Descrybe has opted out. That covers both the vendor's own models and the provider underneath.
It is policy rather than contract: the terms of service only disclose that third party AI models process primary law and carry no matching term. The privacy policy, updated 29 September 2025, was also read and is silent on training.
How long does the product keep what a lawyer typed, and can that be set to zero?
The customer controls the retention window, by product configuration or by contractual instruction, but zero retention is not stated as available.
Retention is described and the user controls it by deletion rather than by a window. The security page (updated 30 July 2026, read 11 September 2026) states that Descrybe retains account and product data needed to operate the service, which may include prompts, research history, generated outputs, metadata and operational logs; that the only upload surface is brief review and uploaded files are not kept once the text is extracted; that no customer facing zero retention mode is offered; and that users can delete their historical entries at any time.
On the provider side it states that OpenAI may retain API abuse monitoring logs for up to 30 days. Recorded as customer controlled with no zero option: deletion on demand is real control over the research history, and the vendor says plainly that zero retention is not available. What the deletion statement does not reach is the operational log and metadata retention, for which no period is stated. Corrected from not addressed on 11 September 2026.
Does retrieval respect the firm’s ethical walls, or can the model read across them?
The product maintains its own permission model, documented, requiring the firm to keep it aligned.
The permission model is documented and it is the simplest one there is: one user, one account, nothing shared. The security page (updated 30 July 2026, read 11 September 2026) states that Descrybe is structured around individual user accounts rather than shared organization workspaces, that saved searches, history and product data are associated only with that user's own account, that access is protected through application and database access controls, and, stated rather than left to inference, that no shared organization workspaces, matter based workspaces or separate file and vault permissioning are offered, on provider managed multi tenant infrastructure.
So walls at matter level do not exist in the product and the vendor says so. A firm that needs them keeps them outside the tool, by giving each user a separate account. Recorded as own model, documented, because the isolation model is published and enforced by the vendor; the summary carries the limit. Corrected from not addressed on 11 September 2026.
If someone subpoenas the vendor for a firm’s data, does the firm hear about it first?
No located term or policy addresses third party requests for customer data.
Not addressed. No clause on government or law enforcement requests, no commitment to notify a user before producing their data, and no transparency report were found. The terms of service cover termination, permitted use and liability, not third party requests. The privacy policy, updated 29 September 2025, covers account information, named third party services and children's data, not this. The security page, updated 30 July 2026, lists legal and compliance obligations among reasons authorized staff may access production data, but says nothing about disclosure to authorities or notice to the user. Checked the terms of service, privacy policy, security page, FAQ and site footer on 11 September 2026.
Where does the law in this product come from, and does the vendor have the right to use it?
Sources are identified without stating the license or rights basis.
Sources are named at dataset level and listed court by court, with no license basis stated. The Harvard Caselaw Access Project is named as a source. The coverage page (data as of 18 July 2026, read 11 September 2026) lists 492 courts and tribunals and 10,969,112 opinions across federal, state, bankruptcy, specialized and agency, and tribal courts, plus statutes, regulations, constitutions, session laws and court rules, with an opinion count and date range for every court and source availability shown where applicable.
Not stated: the license or public domain basis for any source, or which source feeds which court. The content is AI summaries of opinions rather than the opinions alone, so provenance raises a second question the vendor does not address: what the summarization step did to the source.
Does the product tell you when the authority it just cited has been overruled?
The vendor computes and surfaces subsequent history itself, with the method described.
The vendor has its own treatment signal. Cytator launched June 2025 in the Legal Research Toolkit, showing treatment flags in the familiar positive, negative, cautionary and neutral convention, with analysis by issue rather than a single verdict per case. It shows forward treatment with binding and nonbinding citing cases, and a backward citator showing how an opinion treated each authority it relied on. Treatment review is also offered to outside assistants through Descrybe Legal Engine.
It is the vendor's own AI citator, not a licensed one such as Shepard's or KeyCite, and no accuracy, coverage or agreement measure against an established citator is published, which a practitioner would want before relying on a negative flag.
What does the product do when the answer is not in the corpus?
The vendor describes refusal or abstention behavior in public materials.
Documented. The vendor says DescrybeLM tells users when it needs more information and flags uncertainty rather than filling gaps with guesses, and that the engine clarifies jurisdiction, posture and key facts before answering an underspecified question. That is an explicit no answer path described as product behavior. The terms add that outputs are not reviewed or validated for accuracy. No published evaluation, transcript or example shows the behavior working, and no threshold or trigger for the uncertainty flag is described.
Does a public court record exist addressing fabricated or hallucinated legal citations in output from this product?
No court order, opinion or disciplinary record addressing fabricated or hallucinated legal citations produced by this product has been located as of the date shown. This is a statement about the public record on that one subject, not a finding about the product, and this signal is not a litigation history.
None located. 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 docket database or court record tracker was searched. This records what the search found and is not a clearance. The vendor's terms disclose that it does not validate outputs, so a defective citation is an acknowledged possibility rather than one the vendor denies.
Has the vendor engaged in public with the ethics opinions its buyers are bound by?
No located public material engages with bar or ethics guidance.
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.
Does the vendor address what happens to the bill when the work takes an hour instead of six?
The product sits inside a lawyer to client fee relationship and no located public material addresses billing, fee or disclosure treatment, with no savings claim published either.
Not addressed. Nothing published covers billing for AI assisted time or a record a practitioner could show a client of what was machine generated. The vendor's own prices, $25 and $50 a month with all features included, are clear, but that is its charge to the subscriber, not a position on the practitioner's billing. No time record, audit record or fee guidance was found. Checked the home page, the FAQ and the terms of service on 29 Aug 2026.
Can a firm get this vendor through a client’s AI clause without a bespoke negotiation?
A current subprocessor or model provider list is published.
A subprocessor and model provider list is published, each entry linked to that provider's own attestation. The security page (updated 30 July 2026, read 11 September 2026) names OpenAI's API as the primary model provider, DigitalOcean for hosting and storage in its New York region, Clerk for authentication and Stripe for payments, with links to each provider's trust portal or security documentation and the certifications each holds; the privacy policy (updated 29 September 2025) adds Google Analytics for site analysis.
That is the list a firm forwards to a client under outside counsel guidelines, and it exists without a sales conversation. What is not published is client facing disclosure material of Descrybe's own, and Descrybe holds no attestation of its own to add to the pack, which it states plainly. Corrected from not addressed on 11 September 2026.
If a judge’s standing order requires an AI disclosure, can the product produce one?
Some elements of the record are available, short of a document level export.
Partial record, strong on sources. 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, a trail at the tool level.
What is missing is the other two elements 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.
Every grade and every signal on this index is drawn from public sources and dated. If a record is wrong, out of date, or missing an artifact the index did not locate, send the source and it will be reviewed and the record redated. Vendors are welcome to submit documentation. Nothing on this index is for sale, including a listing, a placement, or a grade.