CloudLex vs ProPlaintiff: how they compare in 2026

Lexee's six AI features drive work a firm would miss, on a case management system that stands without them. They cover voice and chat intake, demand package drafting, medical summaries and chronologies, questions about a matter, a client facing website assistant, and voice to task. The medical chronology and the demand draft are core plaintiff work, and the AI drives both, reading from and writing back to the case file. Underneath is a complete practice management system: intake, matters, documents, calendaring, client communication, expenses, settlement calculation, analytics, electronic signature and archiving. The agreement treats the AI Tools as one part of the Services, and Lexee is marketed as one of four pillars alongside the platform, paralegal services and publishing.

The product is AI native by construction and says so. The vendor states it is built from the ground up as an AI native platform rather than layering AI onto outdated software, and the product bears that out. Every named feature is generative or analytical rather than a form with a model attached: demand letters, medical chronologies, document generation, summaries, review, media analysis, case analysis and a conversational assistant. The agentic framing applies to case management itself rather than to one document type, so the model layer is the operating logic of the system rather than one capability within it. Remove the models and what remains is a file store with a calendar. The vendor also calls the AI a force multiplier and digital assistant handling repetitive, data heavy work, which is a centrality claim and a disclaimer of substitution at once.

Grounding is claimed in general terms, the contract is candid about limits, and nothing is measured. Chat answers are described as grounded in the case's documents and notes, and medical summaries are written back into the case file. Nothing says whether a summary or demand paragraph links back to the source record. The agreement states that outputs may not reflect current legal standards, may get people, entities, places or facts wrong, and may not fit the case, and it disclaims accuracy. No accuracy figure, test set, evaluation or error rate is published on the Lexee AI page, the platform page, the pricing page or either terms of service.

Grounding is claimed clearly and nothing behind it is published. The vendor says demand letters cite specific medical records, bills and reports automatically, and that smart citations pull from the firm's uploaded files, which describes citation to the case file rather than invented sources. Its own practitioner guidance says good platforms log what data was used, what was generated and who approved it, and that this is the firm's defensibility. Nothing measures or specifies its own implementation. There is no accuracy figure, extraction error rate or evaluation, and no description of what smart citations point to or how a user checks one. Nothing says how the product behaves when records are incomplete or provider names are inconsistent, a failure mode the vendor's own content says is where platforms fall apart. The citation mechanism is asserted rather than shown: no page level links, hyperlinked navigation or table of authorities is described.

CloudLex publishes its human review as a workflow. Its four step process for medical record summaries ends with Human Review: an attorney or paralegal reviews, edits and approves the output. Demand drafts are described as designed for staff review, and medical summaries as paralegal reviewed. CloudLex's FAQ says human review stays central to every workflow. The agreement requires outputs to be independently reviewed and validated, by a qualified attorney where applicable, before use in proceedings or client communications. The client facing website assistant verifies a caller's identity before releasing case details, under the firm's control. What is missing is the failure path: no abstention, confidence signal or threshold for declining, and no route to report a wrong output.

The oversight position is stated as a rule rather than as reassurance. The vendor's guidance says attorney oversight is non negotiable, that AI drafts and attorneys approve, and that no compliant workflow skips that step. It says the firm owns the output and that a good platform logs what data was used, what was generated and who approved it. The product ships permission management and system activity logs, the mechanisms an approval record would run on. The vendor also disclaims substitution, describing the AI as a force multiplier that frees paralegals and attorneys for client communication, negotiation and trial preparation. The rule is published as advice to the profession, not as a specification of this product. Nothing says ProPlaintiff enforces an approval step, whether any output can be sent without one, what the activity log captures, or whether the platform produces the approval record the vendor says firms need.

Customers are named with their firms and titles: a partner at Pollack, Pollack, Issac & DeCiccio, a senior litigation attorney at Mitchell and Danoff, a founding partner at Rasmussen Injury Law, a trial lawyer at Ikhilov & Associates, attorneys at Glugeth & Pierguidi and the Law Office of Daniel Horowitz, and a paralegal at Abrams & Associates. The one figure is about migration: a firm moved over 300 cases without issue. Nothing measures outcomes such as time saved or intake conversion, and no quote is dated. Client Stories and Client Reviews pages are also published.

Nothing is published. No customer or firm is named anywhere in vendor material, and there is no case study, usage figure, time saving or outcome claim, funding announcement, independent analyst placement or award. A G2 listing exists, carrying a vendor supplied product description rather than verified customer outcomes. The vendor publishes a substantial practitioner content library, and none of it contains a customer result. Checked the home page, the demand letters page, the case management page, the blog library, the G2 listing and the site navigation. This is a documented absence on those pages, not a judgment about adoption, which public material cannot show.

Confidentiality and training carry substantive commitments from CloudLex. The customer owns its data exclusively, and CloudLex may access it only to provide the service, fix problems, comply with the law, or at the customer's request. Safeguards include preventing access by CloudLex staff outside narrow exceptions. CloudLex will not use customer data for training other than deidentified Aggregated Data, and will not let its vendors or subcontractors do so. The AI section commits to keeping client data confidential under privacy laws and bar ethics rules. Two points cut against it: the agreement says customer data may travel unencrypted, and nothing describes separation between matters within a firm.

Basic protections are stated with one useful specific, and there is no privilege position. Published: HIPAA compliance with industry standard safeguards, encryption, storage in the United States, and access limited to authorized users at the firm, supported in the product by permission management, system activity logs and secure case sharing. Naming United States storage is a concrete commitment. Against that, no SOC 2 or other independent attestation is published. No business associate agreement is offered, although the vendor's own guidance tells readers to choose HIPAA compliant platforms under a BAA. Attorney client privilege and work product are not addressed. The compliance posture is self declared, supported by product features.

The professional responsibility position is complete, and it reaches the consumer. CloudLex is not a law firm, cannot give legal advice, and its AI creates no attorney client relationship. Use is limited to subscribers based in the United States. Outputs must be independently reviewed and validated, by a qualified attorney where applicable, before use in proceedings or client communications. The firm is responsible for its fiduciary, record keeping and trust accounting duties, competent representation, deadlines, and getting clients' consent to the service. Users are encouraged to disclose AI use to clients, and the firm is responsible for what the chatbot says to clients and prospects. The separate chatbot terms tell claimants directly that Lexee AI cannot give legal advice. What is missing: no bar authority is named.

The vendor publishes a real professional responsibility position, in its practitioner guidance. It says ABA Model Rules on competence and supervision apply to AI assisted drafting, that the firm owns the output, and that attorney oversight is non negotiable and no compliant workflow skips the approval step. It says firms must verify how damages language, statutory references and negligence framing vary by state. It also disclaims substitution, saying the AI does not replace human judgment and calling it a force multiplier. Naming the rules that apply, who carries the obligation and that jurisdictional variation is the user's responsibility goes beyond a disclaimer. It is published as market education in a blog library, not as a product commitment or a term. Nothing binds the vendor to it, it is not surfaced in the product or any agreement located, and no bar authority beyond the ABA Model Rules is engaged.

Responsible AI principles sit in CloudLex's contract, with nothing published behind them. The agreement's Transparency and Responsible Use section says the AI Tools follow principles of transparency, fairness and human oversight, and encourages users to disclose AI use to clients. Marketing says Lexee is continuously trained to stay reliable and responsible, and is designed to support human judgment. Missing from the Lexee AI page, the platform page and both terms of service: an owner of model behavior, testing before release, any published result, model documentation and bias measurement. That gap matters because the AI talks directly to injured members of the public through intake and the client assistant.

Nothing is published on how the models are governed, evaluated or monitored. There is no AI policy, model card, bias or fairness testing, evaluation method, accuracy monitoring, drift statement, named governance body or ISO 42001. The vendor's content says accuracy of AI medical record analysis has improved for platforms trained on clinical and legal documents, which is a claim about the field rather than a disclosure about this system. The risk is specific here. By the vendor's own description, the platform generates demand letters that anchor damages, including a justified pain and suffering valuation, and any systematic tendency in how that valuation is framed falls on claimants.

Deletion deadlines are real, and for the rest CloudLex leans on Azure. After cancellation a customer has 30 days to export data, and a further 30 days on request for a one time migration fee. CloudLex securely overwrites or deletes customer data 90 days after termination. Its access to customer data is limited to four named cases, and safeguards cover access by CloudLex staff. The security page and FAQ describe AES-256 at rest, TLS 1.2 or above in transit, key rotation, multi factor authentication, role based access, backups and monitoring, attributed to Azure's framework. What is missing: no subprocessor list, and no incident response or breach notice practice. The agreement also says customer data may travel unencrypted across connecting networks and disclaims liability for deleting or failing to store data, and the privacy policy disclaims the unnamed infrastructure provider's failings.

Protection is stated with one location specific, and use is not addressed. Published: data is encrypted, stored in the United States, accessible only to authorized users at the firm, with HIPAA compliance and industry standard safeguards. United States storage is a real commitment and the concrete element here. Not located: any statement on whether uploaded medical records, case files, generated chronologies or demand drafts are used to train or improve models, any retention period, any deletion right at matter close, and anything on what happens to saved templates and preferred language, which by the vendor's own description build up a firm's drafting patterns over time. No exercisable control such as manual deletion and no independent attestation covering the handling claims is published.

CloudLex disclaims the exposure its AI creates and gives nothing back. Damages for all claims are capped at the lower of the last paid invoice or two thousand five hundred dollars, so a monthly customer is capped near one month's fee. No carve out is named for confidentiality, security breach, gross negligence or willful misconduct. The service is provided as is, with every warranty disclaimed, and AI output carries its own disclaimer with no liability for errors or reliance. The only indemnity runs from the customer to CloudLex, covering the firm's own negligence, including malpractice, and any client or prospect interaction through the chatbot or other AI Tools. There is no vendor indemnity, insurance position or service credit. Disputes go to AAA arbitration in New York before three arbitrators, with legal fee recovery capped at twenty thousand dollars.

No published position is published. Nothing covers liability for AI output, warranties, service levels or remedies. The vendor's own guidance says the firm owns the output. That allocates responsibility to the customer and is the closest thing to a liability position in public material, but it is advice to the profession, not a contract term. Nothing addresses what happens when a generated demand misstates a treatment, a citation points to the wrong record, or the AI calendar produces a wrong deadline. On a case management platform, that last one is a malpractice exposure, not a drafting one.

Named connections cover the tools a small firm uses. Microsoft 365 has its own product page, and there is a Word connector and Gmail and Outlook email management that attach correspondence to the matter. A Legal App Integrator, electronic signature, iOS and Android apps and a referral engine for sending cases to other firms are offered. The agreement covers third party modules, each under its own addendum and charged separately. What is missing: no API or developer documentation is published, and no document management or electronic billing system is named.

No integration is available, and the vendor says so. Published material describes API integration as planned rather than delivered, and data migration is offered, which is a one time import rather than an integration. No case management, document management or records retrieval integration is named. The reason is structural: the product is itself the case management platform, competing with Filevine, Litify, Clio and CasePeer rather than connecting to them. That position is coherent, and it leaves a firm with no ongoing route to move data between this system and anything else.

Delivery is in the cloud on Microsoft Azure, which CloudLex names on its pricing page, with round the clock monitoring and 256 bit encryption. The agreement limits use to subscribers based in the United States. No region or data center, selectable residency option, tenancy model, or location for AI processing is stated. Two provisions pull the wrong way. The agreement says customer data may travel unencrypted across connecting networks, and the privacy policy says servers sit with an unnamed third party provider whose security CloudLex does not guarantee.

A real residency statement has nothing behind it, and a deployment option is named without detail. The vendor says data is stored securely in the United States, a jurisdictional commitment rather than a gesture at secure cloud hosting. Custom deployment is named as available for firm wide automation supporting hundreds of attorneys, which implies something beyond standard shared hosting but is not described. No hosting provider, region or availability zone is named, nothing says whether custom deployment means dedicated infrastructure, a private instance or configuration, and no data processing description is given. A country is the start of a residency answer, not the whole of it.

Every certification CloudLex names belongs to Microsoft, and it says so. CloudLex's FAQ says it is built on Azure and inherits Azure's SOC 1, SOC 2 Type II and ISO 27001 certifications and its audited controls, including penetration testing, geo redundant data centers and encryption. The security page adds compliance with HIPAA, IRS and AICPA standards. An infrastructure provider's certification covers the provider's controls, not the application on top. Nothing published claims CloudLex itself has been audited: no CloudLex SOC report, ISO certificate, auditor, scope, application penetration test, trust portal or route to request a report is published. HIPAA compliance is the one claim made in CloudLex's own name, without an attestation.

No independent attestation of any kind is published, and the vendor's own guidance shows it knows the standard. Located: a HIPAA compliance claim with industry standard safeguards, encryption, United States storage, permission management, system activity logs and a stated secure, HIPAA compliant workflow. Not located: any SOC 2 report, ISO certification, named auditor, examination scope or date, business associate agreement, trust center, security page or route to request documentation. The artifact is absent, not gated. The vendor's own advice tells buyers to look for SOC 2 Type II, HIPAA compliance and GDPR adherence, and to protect privileged communications by choosing HIPAA compliant platforms under a BAA. On its own criteria, this vendor would not clear the bar it sets for others.

Nothing published says what runs the AI. No model, provider, version or inference location is named on the Lexee AI page, its six feature descriptions, the platform page, the pricing page or either terms of service. The agreement defines AI Tools by CloudLex's own branding, and marketing calls Lexee a PI trained AI without saying how. The agreement mentions third party providers in a general integration context, naming Google, Facebook, Apple, Microsoft and Amazon as examples, and its training clause binds unnamed vendors and subcontractors. No notice of model changes is promised.

Nothing located. No foundation model provider, model family or version is named, no distinction is drawn between proprietary and third party models, and no subprocessor list was found. The vendor's own practitioner content discusses general assistants including named commercial models and advises that firms require approved confidentiality controls before those touch case files, which demonstrates awareness that the identity of the model matters and stops short of identifying its own. For a platform processing protected health information under a claimed HIPAA posture, every processor touching that data is a question a firm's diligence asks and none can be answered from public material.

No price is published at any level, and CloudLex says so. Its pricing page states it does not publish standard per user pricing, that pricing is customized by package, users, intake and case management needs, storage, integrations, AI and optional services, and that third party price figures are estimates only. There is no figure, rate, unit or package structure to price. The agreement gives billing mechanics instead: monthly or yearly billing in advance, automatic renewal at the original price, a ten day window to opt out after a renewal notice, no refunds for partial months or downgrades, a credit card on file, extra charges above subscribed users or matters, and separate charges plus a service fee for third party modules.

No pricing is published at any level: no price, range, tier structure or unit of charge. Nothing shows how custom deployment for hundreds of attorneys is priced against a solo practitioner, although both are named as buyers. No independent estimate is published either. In contingency work, case costs are advanced against a claimant's recovery, so undisclosed cost is ultimately cost to an injured person.

CloudLex documents its coverage and states one hard limit. Nine practice areas have their own pages: personal injury, slip and fall, product liability, mass torts, workers' compensation, social security disability, motor vehicle accidents, wrongful death and medical malpractice. It is built only for plaintiff personal injury firms. The agreement limits use to subscribers based in the United States. Paralegal services cover no fault litigation processing, court calendar management and bilingual intake. What is missing: no firm size, matter volume or state level coverage for deadlines and calendaring is stated.

Functional breadth is wide, practice scope is narrow and plainly stated, and the coverage that matters for a case management platform is not described. Functions run from intake through litigation: case management, calendaring and deadlines, document generation and review, media analysis, chronologies, demands and analytics in one system. The practice is plaintiff personal injury, with solo attorneys, paralegals and firms of hundreds of attorneys named as buyers. Nothing is described at the level a buyer checks. No jurisdictional coverage is stated, although the vendor's own guidance warns that damages language, statutory references and negligence framing vary by state. Nothing says which record types or formats the medical analysis handles, or whether areas such as mass tort or medical malpractice are covered. For the firm's system of record, coverage claims need boundaries, and none are given.