DecoverAI vs Discernis: how they compare in 2026
DecoverAI and Discernis both run their AI across every document in a collection to call responsiveness and privilege, explain each call, and hand the result to a reviewer to confirm or override. DecoverAI's lead is written into its terms. They bar training on customer data without written consent, state that nothing in the agreement waives privilege, and name ABA Model Rules on competence and confidentiality. They also commit to breach notice within 72 hours, and a dated subprocessor list names each AI provider. Discernis states its data commitments in an FAQ rather than its terms, which date from May 2024 and carry no training or confidentiality clause. Discernis answers on where the AI runs. It builds and hosts its own models, calls no outside AI service, and runs in its own cloud, a private cloud, on the customer's premises or air gapped, with the hardware for a local installation published. DecoverAI's inference goes to OpenAI, Anthropic and Google, partly through a routing layer that reaches further models.
Which one fits
Choose DecoverAI if
- You want your data commitments in the agreement you sign. DecoverAI's terms bar training on customer data without your prior written consent, state that nothing in the agreement waives privilege, restrict its staff from reviewing your data except for authorized support, and delete data 30 days after the term ends, with written certification on request.
- You want to price a matter before you start. DecoverAI publishes $60 per gigabyte per month with no seat fees, a second unit at two cents per document, a worked range of $300 to $1,200 for a typical matter, a cost estimator and self serve signup.
- Your documents live in iManage, Clio or Microsoft 365. DecoverAI names customer authorized connections to iManage, Clio, SharePoint, OneDrive, Microsoft 365, Google Drive, Gmail, Dropbox, Box and customer owned Amazon S3, each configured by the customer, and produces privilege logs and Bates numbering to firm conventions.
Choose Discernis if
- You cannot send a collection to an outside AI company. Discernis states that it builds and hosts its own models, that no third party or commercial AI service is called at any point, and that inference runs inside whichever environment you choose.
- Your documents cannot leave your building. Discernis runs in its own cloud on US based Azure by default, in a private cloud, on any major cloud or Kubernetes cluster, or on your premises, including air gapped, and publishes the hardware a local installation handling a terabyte a month needs: about 32 CPUs, 128 GB of RAM, 2 TB of storage and eight H100 GPUs.
- You want to pay per document with no charge for storage or users. Discernis prices Discovery per document, on matter or annual terms, with no per gigabyte, per user, storage, ingestion or export charges, and allows cancellation at any time. The rate itself is quoted before a matter starts.
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.
Classification for responsiveness, privilege and confidentiality is what compresses the review, and every downstream artifact depends on it: the privilege log is generated from the AI's privilege calls, redaction is driven by automated detection with attorney override, and the headline claim is 80 percent less review because the model triages every document. The company markets a webinar on how it post trains an LLM to run document review at under five cents a document. Without the models, document storage with Bates numbering remains.
Discernis Discovery takes the questions from a review protocol and has the company's own models read every document in the collection, assess its responsiveness, prioritize it, flag likely privileged communications and explain each call. Discernis Investigations reads the same kind of collection against chosen topics to build timelines and map parties. The Discernis FAQ states that the company builds and hosts its own models rather than calling an outside AI service, and the pricing, deployment options and review workflow are all organized around that model output.
Citation Accuracy and Hallucination Disclosure
Whether the vendor publishes measured accuracy on citations and assertions, grounds output to primary sources, and says plainly what its system does when it does not know. Legal has a documented public record of fabricated citations reaching filed briefs, so an untested claim of accuracy is not evidence.
The published subprocessor list names Parallel Web Systems for web search grounding of AI answers and citations, Cohere for reranking search results, and UniCourt for case law and litigation data feeding AI research features, so a reader can see what answers rest on and who supplies each part. The agreement names failure modes: clause 13.5 refers to incorrect, inaccurate or hallucinated AI output, including case citations and regulatory references. No accuracy figure, test set or evaluation is published. A white paper on defending AI review is published.
Each tag is tied to the document it describes and carries the model's explanation, which reviewers check in a side pane, and results export with a score and an explanation for every responsive document. The product reads every document in the collection rather than retrieving a sample. The Discernis FAQ states about 99 percent accuracy, with recall above 95 percent and precision above 90 percent, and says agreement between annotators is indistinguishable from top reviewers. The Discovery product overview gives about 95 percent or more on relevance classification, about 90 percent precision and about 95 percent recall. Neither set of figures describes a test set, matter type, sample size or date, and no failure modes are named.
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.
The product position is that every classification is reviewable and every redaction overridable, with redactions burned without touching originals so the underlying document survives an override. The commitment is contractual as well as marketing: clause 2.5 makes the customer solely responsible for supervising and verifying all AI generated outputs before professional reliance, and clause 13.5 requires independent review and verification by a licensed attorney. Clause 4.3 keeps access and activity logs for at least twelve months and makes them available on written request. No threshold is published at which the system stops or escalates, no confidence signal is described as visible to the reviewer, and nothing states what happens after a misclassification is found beyond the customer's own override.
The models run first pass review across the whole collection on their own. The Discernis FAQ states that reviewers can override or correct any AI suggestion, that a quality control workflow validates the AI tags, and that a side pane shows each AI answer for human validation. An audit log records extraction, editing and validation by date, time and person and cannot be edited by users. The home page presents the explanations as the basis for a second pass human review. Nothing published sets a threshold at which a document is routed to a person rather than tagged, states what the output may not be relied on for without review, or covers what happens when a tag proves wrong.
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.
Six case studies are published, three naming the firm: Clayton Trial Lawyers, with a 15.4 million dollar jury verdict after a five week trial; Schaff Law Group, with more than 100 hours saved on case preparation; and Gregor Wynne Arney PLLC, processing more than a million documents in a 35 million dollar healthcare fraud and anti kickback investigation. Three more give figures without a firm name, including 30,000 documents produced in three days saving 147,000 dollars and 25 days, a remediation of more than 360,000 documents fixing six of six defects under federal scrutiny, and a construction defect matter covering a terabyte across three buildings. The figures are tied to the named firms. No case study is dated, and no method appears on the summary cards.
The home page claims 90 percent faster review, 70 percent lower cost and about 99 percent accuracy. The law firm page claims a 40 to 50 percent cost reduction. The funding announcement of 25 August 2026 says the product is in use at Am Law firms on active matters. No customer is named, no matter or date is attached to any figure, and no method is given for how the savings were measured. No named case study is published on the case studies page, in the newsroom or on the blog.
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 7.3 treats all Customer Data as Confidential Information and says nothing in the agreement waives any legal privilege or protection over it. Clause 10.4 gives privileged Customer Data the highest level of protection, instructs staff not to review Customer Data except as strictly needed for technical support and only with the customer's prior authorization, and binds everyone with possible access to written confidentiality. Clause 7.9 prohibits training without prior written consent. Separation is documented: single tenant deployment, private VPC, deployment in the customer's own VPC, least privilege role based access, and processing in isolated environments destroyed after each job. Clause 7.10 sets retention and deletion, with irreversible deletion and written certification available. Nine AI and machine learning subprocessors are named with their role and country.
The Discernis FAQ says no data is retained for training and that data is removed from all systems within 30 days of a user deleting it. On the home page, Discernis says it builds and hosts its own models, so documents are never sent to outside AI companies, and that the platform can run on the customer's own premises or air gapped. The published Terms of Use, last updated 1 May 2024, carry none of this. They contain no confidentiality or training term and state that the company may access, store, process and use any information and personal data a user provides. They also say the Services are not tailored to comply with HIPAA, while the Discernis FAQ states HIPAA compliance. Nothing is published on privilege or work product handling, or on separation between matters or users.
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 2.5 says the Services are productivity tools and not legal advice, and names the duties engaged: competence under ABA Model Rule 1.1, confidentiality under ABA Model Rule 1.6, and supervision. Clause 11.4 says the Services do not replace professional legal judgment and no generated content is legal advice. Clause 13.5 requires a licensed attorney to review and verify output before relying on it and puts the risk of unverified reliance on the customer. Clause 4.4 requires the customer to warrant it has any client consent needed under bar rules or ethics opinions on cloud legal technology, and clause 2.3 bars use that breaks professional responsibility rules. No jurisdiction limit on substantive coverage is stated, and the product works on the customer's own documents; the authorities named are American and California law governs.
Nothing states what the product is and is not, that its tags and explanations are not legal advice or a substitute for a lawyer's judgment, or who may use it, and the Terms of Use carry no advice disclaimer. The Discernis blog advises in house counsel to keep meaningful, documented oversight of AI findings and calls private deployment a professional responsibility requirement. The Investigations product is marketed for HR complaints and staff activity reviews as well as legal matters, and no audience limit is stated for it.
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.
No AI governance framework is published: no one is named as accountable for model behavior, no evaluation before release is described, no test results are published, and nothing addresses bias or uneven output, although the product's core job is classifying documents as responsive or privileged. SOC 2 Type II covers security, availability and confidentiality, controls are said to be monitored continuously and shown live in a trust center, and a white paper on what courts expect from AI assisted review is published. Security certification is a different subject from AI governance.
The product explains every tag it makes, and the Discernis FAQ says the company measures its models against human reviewers, including agreement between annotators. No governance framework, accountable owner, testing regime before release or bias finding is published. The Investigations product is marketed for HR discrimination complaints and for reviewing staff activity for signs of burnout and disengagement after layoffs, and no statement addresses uneven output across groups of people.
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 a stated period: clause 7.10 retains Customer Data for the subscription term plus thirty days for export, then irreversibly deletes or destroys it, with written certification of deletion on request, and clause 7.7 lets the customer request removal at any time with a commitment to process it within thirty days. Access control under clause 7.4 covers TLS 1.2 or higher in transit, AES-256 or equivalent at rest, role based access and least privilege for DecoverAI personnel, and regular penetration testing and vulnerability scanning by independent third parties. Incident practice is contractual: clause 3.5 commits to notifying the customer of any confirmed security breach affecting Customer Data within seventy two hours of confirmation. The subprocessor list is public, dated and versioned, naming every provider, its function and its country, with thirty days' prior written notice before any addition and a subscribable list for change notices. Clause 4.3 retains access logs for at least twelve months and releases them on request, and clause 7.11 gives an annual customer audit right.
The Discernis FAQ says nothing is retained for training and that data is removed from all systems within 30 days of a user deleting it. Hosting defaults to a US based Azure cloud, with other regions or clouds available on request, and the platform can run on premises or air gapped so documents never leave the customer. An audit log records every extraction, edit and validation by person and time and cannot be altered by users. No named subprocessor list or incident or breach notification practice is published. The Privacy Notice, last updated 1 May 2024, covers website visitors and account details and does not address documents loaded for review. A trust center is linked from the FAQ.
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 12.1 indemnifies against patent, copyright, trademark and trade secret claims, with five exclusions at 12.3 and remedies at 12.2. Clause 13.2 caps each party at twelve months of fees, and clause 13.3 lifts the cap for breach of confidentiality, gross negligence or willful misconduct and for DecoverAI's indemnity, leaving confidentiality breach uncapped. Clause 11.2 gives six warranties, including that the Services will perform materially as documented and that DecoverAI maintains SOC 2 Type II certification, turning a marketing claim into a contract promise. Clause 14.1 commits to 99.9 percent uptime with service credits. No insurance position is published. Clause 13.5 disclaims liability under any theory for incorrect, inaccurate or hallucinated AI output, naming case citations and regulatory references, and puts all reliance risk on the customer.
The Terms of Use, last updated 1 May 2024, provide the Services as is, with all warranties disclaimed, including any warranty on the accuracy or completeness of content. The company's liability is capped at the amount paid in the six months before the claim arose. The user indemnifies the company, with no indemnity running the other way. The Discovery product overview carries the heading Remarkable Accuracy, Guaranteed. No guarantee terms are published, and the terms disclaim accuracy.
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.
The subprocessor page lists customer authorized connections: Microsoft SharePoint, OneDrive and Microsoft 365 through the Graph API, Google Drive and Gmail, Dropbox, Box, customer owned Amazon S3 buckets, and two legal systems, iManage and Clio. It says DecoverAI reaches these sources only at the customer's direction and under the customer's own agreement with each provider, and that the customer configures the connections. Appendix A of the agreement lists connectors to document management systems, email platforms and cloud storage as a contracted service element set in the Order Form.
No integration with a review platform, document management system or matter system is named. Data moves in and out by file. The Discernis FAQ says the platform accepts load files.dat files and native files, and exports load files.dat files and a CSV listing responsive documents with their scores and explanations, with no ingestion or export fees. The format matches the load file exchange that review platforms already use. Work moves between Discernis and other systems by export and import rather than through a connection.
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.
Three options are published: default multitenant, single tenant with dedicated infrastructure, and deployment in the customer's own VPC for full isolation, all with AES-256 at rest and TLS 1.2 or higher in transit, and processing in isolated environments destroyed when each job finishes. The subprocessor list gives a country for every provider, placing AWS hosting and storage in the United States, optical character recognition by Mistral AI in France, and reranking by Cohere in Canada, which separates where processing happens from where data is stored. No residency offering is published: no region can be chosen, and nothing lets a customer require data to stay in a given jurisdiction.
The product overview names three options, Discernis Cloud, Discernis Private Cloud and on premises, and the Discernis home page states full functionality in each. The Discernis FAQ adds air gapped installation and deployment on any Kubernetes cluster and any major cloud provider, naming Google Cloud, AWS and Azure. Hosted data sits in Azure in the United States by default, and the company will deploy to any region or other cloud on request. The funding announcement says all inference runs inside the environment the customer chooses, with no third party or commercial AI service called. A local deployment handling a terabyte of data a month needs about 32 CPUs, 128 GB of RAM, 2 TB of storage and eight H100 GPUs, according to the FAQ. Cloud offerings scale with demand, while local throughput depends on the compute allocated.
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.
SOC 2 Type II is claimed for the security, availability and confidentiality criteria, verified over an observation period rather than at one point. HIPAA compliance is claimed, with a Business Associate Agreement on request. Clause 11.2(f) warrants that DecoverAI keeps SOC 2 Type II certification, and clause 7.11 gives an annual audit right that DecoverAI may meet with its latest SOC 2 Type II report and a third party penetration test summary. A trust center at trust.decover.ai is said to show control monitoring live. No auditor, report date or observation period is published, and the full report is available only under NDA. GDPR is described as certification in progress.
Discernis holds no independent security attestation. The Discernis FAQ states compliance with HIPAA, which has no certification scheme of its own, and says the company is working to finalize ISO 27001 and SOC 2. Neither is claimed as achieved, and no auditor, scope or date is given. The published Terms of Use say the Services are not tailored to comply with HIPAA and may not be used where HIPAA applies. A trust center is linked from the FAQ.
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.
The subprocessor list, dated August 2026 and versioned, names nine AI and machine learning providers with their role and country: OpenAI, Anthropic and Google through its Gemini API for inference in document analysis, classification and question answering, all in the United States; OpenRouter as a routing layer to more models; Mistral AI in France for optical character recognition; Cohere in Canada for reranking; E2B for isolated code execution in agent workflows; Parallel Web Systems for web search grounding; and UniCourt for case law and litigation data. AWS is named for hosting, storage, databases, search, authentication and content delivery. Additions require thirty days' prior written notice, with a notification list, repeated in clause 7.5 of the agreement. Gemini is named as a model family and the rest only by provider, and the OpenRouter layer means the model behind a given request is not disclosed by design.
The Discernis FAQ states that the company builds and hosts its own models, built for discovery. The funding announcement says no third party or commercial AI service is called at any point, with inference running inside the deployment the customer chooses and hosted by default on a US based Azure cloud. The architecture, and any base model the proprietary models are built from, is not identified, and no commitment to notify customers when the models change 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.
The rate is 60 US dollars per gigabyte per month, described as all in, with no seat fees, enterprise tier or contract, and the page lists what it covers: AI classification for responsiveness, privilege and confidentiality, redaction detection with attorney override, an automatic privilege log spreadsheet, Bates numbering to firm conventions, a full audit trail for regulatory productions, and SOC 2 compliant encrypted hosting. A second unit is two cents per document, with an example that most teams process five to twenty gigabytes per matter for 300 to 1,200 dollars all in, and a cost estimator and self serve signup are offered. The marketing says no contracts and no seat fees, while the agreement provides for Order Forms with user seats, capacity limits, overage rates, annual invoicing in advance, sixty days' notice to stop renewal, and fees due to the end of the term. Both are published.
The Discovery page states pricing per document, with no charges per gigabyte, per user or for storage. The FAQ adds no ingestion or export fees. The pricing page offers matter and annual pricing options, fast setup and cancellation at any time, and the Terms of Use describe a monthly subscription that renews automatically. A blog post explains that the price per document is meant to be quoted before a matter starts. No rate is published. The law firm page describes the offer as having no charges per document, which conflicts with the unit stated on the Discovery page and in the pricing blog post.
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.
Four stages of litigation have their own pages: ediscovery, early case assessment, internal investigations covering cyber breaches, whistleblower actions, subpoenas and regulatory inquiries, and trial preparation. Case studies show the matter types: commercial litigation, personal injury, white collar crime, a tax credit investigation, a federal production remediation and a multiparty construction defect matter. Buyers are law firms and in house legal teams, and HIPAA is positioned as the baseline for personal injury, medical malpractice and healthcare investigations. Clause 2.4 addresses US Government use rights under FAR and DFAR. No firm size, matter volume range, or excluded matter or data type is stated.
Separate pages address in house legal departments, law firms and legal service providers. The uses are first pass responsiveness and privilege review in litigation discovery, and fact finding in internal investigations, compliance reviews and case strategy. No document types, matter sizes or practice areas are named as unsupported. The Investigations examples reach beyond legal work into HR, manufacturing and operational risk.
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?
Clause 7.9 of the published agreement, headed No Training on Customer Data, provides that DecoverAI will not use Customer Data or Inputs to train, fine tune or improve any artificial intelligence or machine learning model without the customer's prior written consent, and adds that its models are trained exclusively on licensed datasets and public information sources. Training is prohibited by default and can occur only where the customer affirmatively agrees in writing, which is opt in rather than an unqualified prohibition.
Clause 7.6 is consistent, licensing Inputs and Customer Data solely to operate the Services, address technical problems and meet legal obligations, and expressly withholding any commercial purpose unrelated to the Services absent prior written consent. DecoverAI's home page states the position more absolutely than the agreement does, as your data never trains our models, without the consent carve out.
The Discernis FAQ answers no to whether any data is retained for training, and Discernis says on its home page that it builds and hosts its own models, so documents are never sent to outside AI companies. The published Terms of Use, last updated 1 May 2024, carry no training term in either direction. They state that the company may access, store, process and use any information and personal data a user provides, without naming training. The commitment sits on a policy page, not in the agreement.
Prompt and Output Retention
How long does the product keep what a lawyer typed, and can that be set to zero?
Clause 7.10 sets a specific period: Customer Data is kept for the subscription term plus thirty days for export, then irreversibly deleted unless law requires otherwise, with written certification of deletion on request. Customers can shorten that by instruction, since clause 7.7 lets them request removal of Customer Data in writing at any time, processed within thirty days. Zero retention is not offered as a setting, though security material says documents are processed in isolated environments destroyed when each job finishes.
The customer controls how long documents stay. The Discernis FAQ says data is completely removed from all systems within 30 days of a user deleting it, and on premises or air gapped deployments keep the collection inside the customer's own environment. A setting that retains nothing is not described. The Privacy Notice keeps account information for as long as an account exists, and addresses website and account data rather than documents under review.
Ethical Walls and Matter Segregation
Does retrieval respect the firm’s ethical walls, or can the model read across them?
DecoverAI maintains its own documented permission and isolation model rather than inheriting a source system's access control at query time. Published detail covers single tenant deployment and private VPC, with dedicated infrastructure or deployment inside the customer's own VPC for organizations requiring full isolation, AES-256 at rest and TLS 1.2 or higher in transit by default, enterprise single sign on through the customer's own identity provider, role based access controls and enforced multifactor authentication.
Clause 7.4 adds role based access and least privilege for DecoverAI personnel, and clause 10.4 restricts personnel from reviewing Customer Data except as strictly necessary for technical support and only with the customer's prior authorization. The customer administers its own Authorized User access under clauses 2.2 and 3.2. No material addresses segregation between separate matters inside one customer account.
No published material addresses ethical walls or separation between matters or users within an account, in the product pages, the FAQ, the Terms of Use, the Privacy Notice or the Discovery product overview. The FAQ describes an audit log of who extracted, edited and validated what, and an on premises deployment leaves access to the system with the customer. Neither is a published permission model at the matter level.
Third Party Request and Subpoena Notice
If someone subpoenas the vendor for a firm’s data, does the firm hear about it first?
Clause 10.3, Compelled Disclosure, commits a party required by law or court order to disclose the other's Confidential Information to give prompt written notice where lawful, help seek a protective order, and disclose only what is legally required. Customer Data is Confidential Information, and clause 10.4 gives privileged information the highest protection, so the notice reaches what a firm cares about most. No transparency report of government or third party requests is published.
The Privacy Notice, last updated 1 May 2024, says information may be processed to comply with legal obligations and respond to legal requests, and the Terms of Use reserve the right to report users to law enforcement. Neither document commits to telling the customer before its data is disclosed, or reserves a discretion to do so. The FAQ presents deployment inside the customer's own environment as keeping data out of reach of outside legal requests, which applies where the platform runs on the customer's premises.
Primary Law Corpus Provenance
Where does the law in this product come from, and does the vendor have the right to use it?
Clause 7.9 states that DecoverAI's AI models are trained exclusively on licensed datasets and public information sources, which characterizes the rights basis without identifying any dataset, publisher or collection. The published subprocessor list separately names UniCourt as the supplier of case law and litigation research data feeding AI assisted legal research features, which identifies a supplier rather than a corpus.
No primary law source, jurisdictional coverage or update cadence is published on the home page, in the terms of service, the subprocessor list or the security material, and the product's principal corpus is the customer's own document set rather than an external legal collection.
The product works on the customer's own document collection, and no body of primary law behind its output is identified in the home page, the product pages, the FAQ or the Discovery product overview.
Good Law Verification
Does the product tell you when the authority it just cited has been overruled?
Nothing on the home page, the product pages in the navigation, the terms of service or the subprocessor list addresses whether authority is checked for later history. The product is an eDiscovery platform whose core work is classification, redaction and production over the customer's own documents, so the question arises only through the AI assisted legal research features referred to in clause 13.5 and supplied with data by UniCourt. No treatment signal, currency check or citator relationship is described anywhere.
The product classifies and explains documents from the customer's own collection rather than citing legal authority. Nothing addresses checking authority for subsequent history.
Refusal and Uncertainty Behavior
What does the product do when the answer is not in the corpus?
Nothing published describes what the product does when it cannot ground an answer. Hallucination is acknowledged squarely in clause 13.5, which disclaims liability for incorrect, inaccurate or hallucinated information generated by the AI features, but acknowledgment of the risk is not a description of behavior. No abstention path, no state for declining to answer and no confidence or grounding score visible to the reviewer is described, and the published control is human override of every classification rather than anything the system does itself.
Every document receives a responsiveness score and an explanation, which export together in a CSV, and reviewers validate the tags in a side pane with a quality control workflow. No path is described in which the system declines to classify a document it cannot assess.
Fabricated Citation Record
Does a public court record exist addressing fabricated or hallucinated legal citations in output from this product?
The AI Hallucination Cases database maintained by Damien Charlotin records no case naming DecoverAI or DecoverHQ, and no court order, opinion or disciplinary record naming the product is on the public record. This is a statement about the public record rather than a finding about the product.
The AI Hallucination Cases database maintained by Damien Charlotin records no case involving Discernis. No court order, opinion or disciplinary record naming the product is on the public record.
Bar Guidance Alignment
Has the vendor engaged in public with the ethics opinions its buyers are bound by?
Clause 2.5 makes the customer responsible for compliance with applicable Rules of Professional Conduct and names three duties with their sources: competence under ABA Model Rule 1.1, confidentiality under ABA Model Rule 1.6, and supervision. Clause 4.4 requires the customer to warrant that uploading client materials does not violate any court order or rule of professional conduct and that it has obtained any client consent required under applicable bar rules or ethics opinions governing cloud based legal technology, which engages the specific ethics question this product raises.
Clause 2.3 prohibits use of the Services in violation of applicable professional responsibility rules and bar regulations. The guidance engaged is that of one jurisdiction, and no mapping of product behavior to named opinions, clause by clause, is published.
A blog post on defensible AI investigations says several bar ethics committees have weighed in on using AI tools with client confidential information, and calls private deployment for sensitive matters a professional responsibility requirement. No ethics opinion is named, and the product pages, FAQ and terms do not engage with professional guidance.
Billing and Fee Posture
Does the vendor address what happens to the bill when the work takes an hour instead of six?
Published claims include 98 percent cost reduction against traditional attorney review, 147,000 dollars and 25 days saved on a production of 30,000 documents, 20,000 dollars saved on a subpoena response, and pricing framed as 300 to 1,200 dollars per matter against weeks of associate hours. A cost estimator tool estimates the customer's own review spend rather than anything disclosable to a client. Nothing addresses how AI assisted work should be billed or disclosed, and although the product generates detailed artifacts for each matter, including privilege logs and audit trails, none is described as a record of AI assisted work for fee purposes.
The home page claims review that is 90 percent faster and 70 percent cheaper than traditional review, and the law firm page claims a 40 to 50 percent cost reduction. A blog post on pricing per document discusses predictable budgets and easier cost conversations with clients. Nothing addresses how AI assisted review is billed or disclosed to the client when the hours fall.
Outside Counsel Guideline Readiness
Can a firm get this vendor through a client’s AI clause without a bespoke negotiation?
A dated, versioned subprocessor page at decover.ai/legal/subprocessors, last updated August 2026, names every subprocessor for Customer Data with role and country. The same page names nine AI and machine learning providers and what each does: inference by OpenAI, Anthropic and Google Gemini, routing by OpenRouter, optical character recognition by Mistral AI, reranking by Cohere, isolated execution by E2B, web search grounding by Parallel Web Systems and case law data by UniCourt.
It says it is the list referenced in the Data Processing Agreement, that each subprocessor is bound by equally protective terms, including Standard Contractual Clauses where they apply, that additions need thirty days' prior written notice, and that a countersigned DPA and SCCs are available on request. The DPA is published at decover.ai/legal/dpa and incorporated by clause 7.2.
No subprocessor or model provider list is published on the home page, product pages, FAQ, Terms of Use or Privacy Notice. A trust center is linked from the FAQ. Discernis states that it uses its own models, calls no third party AI service, retains no data for training and hosts on a US based Azure cloud by default.
Court Disclosure Support
If a judge’s standing order requires an AI disclosure, can the product produce one?
Published: an automatic privilege log as a spreadsheet, Bates numbering to firm conventions, a full audit trail for regulatory productions and defensibility, and access and activity logs kept at least twelve months under clause 4.3 and released on written request. A white paper on what courts expect from AI assisted review and how to document it is published. Nothing says the model behind a classification is recorded or disclosed, and the OpenRouter routing layer means the specific model handling a document is not shown.
Exports include a CSV listing responsive documents with their scores and the model's explanations, and the audit log records extraction, editing and validation by date, time and person. No record per document combining the model used, the basis for the call and the human verification is described as an export, and no disclosure guidance or template is published.
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.
- Primary Law Corpus Provenance
- Good Law Verification
Product updates
The latest logged changes from each vendor, newest first, each linked to its source.
DecoverAI
-
Product / capability
DecoverAI, built for litigation review, now covers transactional work.
Source
In summary
DecoverAI
DecoverAI is an eDiscovery platform from DecoverHQ, Inc. of San Mateo, California, that takes a document set from upload to production. It classifies documents for responsiveness, privilege and confidentiality, redacts with attorney override, and generates privilege logs, Bates numbering and an audit trail, plus chronologies and evidence analysis. According to the AI Legal Index, its terms bar training on customer data without written consent, state that nothing in the agreement waives privilege, and name ABA Model Rules 1.1 and 1.6. Its subprocessor list names each AI provider it uses. Pricing is published at $60 per gigabyte per month. Seats are not charged, so adding colleagues costs nothing extra.
Discernis
Discernis is an AI document review platform from Discernis, Inc. of New York, sold to law firms, corporate legal departments and legal service providers. Its Discovery product runs the company's own models across every document against a review protocol written in plain English. It scores responsiveness, flags likely privilege and explains each call. Discernis Investigations reconstructs timelines and maps parties across a collection. According to the AI Legal Index, Discernis runs in its own cloud, a private cloud, on premises or air gapped, and calls no outside AI service. Its data commitments sit in an FAQ rather than its Terms of Use, which date from May 2024.
Questions buyers ask
DecoverAI vs Discernis: which is better for AI document review?
DecoverAI's agreement, subprocessor list and pricing say far more: its terms bar training without written consent and address privilege and professional duties, and it publishes a price per gigabyte. Discernis leads on one thing that may decide it for some buyers: it runs its own models and can be installed on premises or air gapped, so documents never reach an outside AI company. DecoverAI's inference runs through OpenAI, Anthropic and Google. From the AI Legal Index, based on each vendor's own published materials as of September 25, 2026. No vendor pays for placement.
Does DecoverAI train on client documents?
Not without written permission. Clause 7.9 of DecoverAI's agreement, headed No Training on Customer Data, provides that it will not use customer data or inputs to train, fine tune or improve any AI model without the customer's prior written consent, and says its models are trained only on licensed datasets and public sources. Its home page states the position more absolutely, that customer data never trains its models. Discernis says in its FAQ that no data is retained for training. From the AI Legal Index, based on each vendor's own published materials as of September 25, 2026. No vendor pays for placement.
Where does Discernis host data, and can it run on premises?
Yes to on premises. Discernis hosts by default in a US based Azure cloud and will deploy to other regions or clouds on request, and it also offers a private cloud, installation on any Kubernetes cluster or major cloud, and on premises or air gapped installation. It states that all inference runs inside the environment the customer chooses. DecoverAI offers multi tenant and single tenant hosting in the United States, and deployment inside the customer's own VPC. From the AI Legal Index, based on each vendor's own published materials as of September 25, 2026. No vendor pays for placement.
Which AI models does DecoverAI use?
DecoverAI's dated subprocessor list names OpenAI, Anthropic and Google's Gemini for inference, OpenRouter as a routing layer to further models, Mistral AI for OCR, Cohere for reranking search results, E2B for sandboxed code execution, Parallel Web Systems for web search grounding and UniCourt for case law data, with each provider's country, and commits to 30 days' notice before any addition. 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.
Two things to weigh. Discernis's published documents disagree with each other. Its FAQ states HIPAA compliance, while its Terms of Use say the service is not tailored to comply with HIPAA and may not be used where it applies. One page says there are no per document charges, while others state per document pricing. The terms are the binding text. DecoverAI's home page says customer data never trains its models, while its agreement allows training with the customer's prior written consent. Discernis links a trust center from its FAQ. Neither vendor reviewed this page.