L

Litem

The models are the product, with no conventional software layer beneath them. The company describes AI workflows and tools across the litigation lifecycle: starting in discovery, assisting through depositions in real time, and offering agentic features for later phases. Every named product line is an inference function rather than a storage or workflow layer. They are analysis of transcripts and testimony, a facts capability over the case record, and an agent for later phases.

Real time help during a live deposition is not something a database does. The alternative is reading transcripts end to end by hand, which the product replaces rather than speeds up, and a customer reports preparation time cut in half. Remove the models and nothing remains that a litigator would license. The agreement fits that reading. It defines third party large language model providers as a term and disclaims warranties specifically for AI generated content, which is a company contracting around model output rather than software.

Accuracy is addressed candidly, but only as a disclaimer, with no measurement. Clause 9.6 of the master services agreement is explicit. No warranty is given on the accuracy, completeness or reliability of AI generated content, the customer is solely responsible for it, and the customer must verify output before use. That is a plain statement in the operative contract. Everything on the other side is absent. No accuracy figure, benchmark or evaluation is published for any feature.

Nothing describes how a summary or extracted fact is tied to the underlying transcript. That matters here because deposition testimony has a natural citation form in page and line numbers, and the vendor's own blog writes about citing transcripts correctly. Nothing addresses hallucination by name. The five solution pages were not opened and may describe grounding.

Oversight is set out in the agreement rather than described in the product, and the allocation is thorough. Clause 9.5 places vetting, output review and privilege preservation on the customer, naming three separate obligations. Clause 9.6 requires AI generated content to be verified before use. Clause 3.3 restricts who may use the system at all: attorney users must be licensed and in good standing in the United States, and paralegal use must not constitute the practice of law.

The identity and competence of the reviewer is therefore contractual, and a firm knows who must check the output and what they answer for. The product side is missing. Nothing published says what runs unattended, where a review checkpoint sits in the interface, what confidence or uncertainty is shown, or how the agentic features for later phases differ in autonomy from the analysis features.

Named firms from the defense bar, volume figures and one attributed outcome. Two customers are named on the vendor's own pages, Ogletree Deakins and Rutan & Tucker, both substantial defense firms. Scale is published as more than fifty Am Law 200 firms, more than 10,000 depositions handled a month and more than 20,000 professionals using the platform. Three people speak on the record, including Kyre Stucklin, Director of Innovation at Rutan & Tucker, and a litigator who reports preparation time cut by half, which ties a figure to a named person.

Method and dates are missing: none of the volume figures has a measurement basis or period, and no case study sets out before and after results. One contractual fact belongs beside any logo strip. Clause 12.1 lets Deposely use customer names and logos, so a logo shows a contract rather than an endorsement.

Training and retention are answered firmly in the agreement; separation and providers are not answered at all. On training, clause 5.4 ends with an express statement that Deposely will not train artificial intelligence models using customer data as training materials, which prohibits the activity in the operative contract. On retention, clause 6.4 requires immediate deletion of all proprietary information, including customer materials, promptly after termination, subject only to legal record keeping.

Clause 5.4 also limits usage data to anonymized collection and aggregated disclosure. The privacy policy does not cut against either. It excludes content uploaded for business customers from its scope, so its language about improving algorithms and AI models reaches prospect and visitor data, not matter material. Privilege itself appears only in clause 9.5, which places its preservation on the customer rather than describing how the platform supports it. Nothing describes separation between customers or between matters, and no model provider is identified.

Three clauses in the contract do the work. Clause 9.4 states that Deposely is solely a technology provider and provides no legal services or advice. Clause 9.5 sets out the professional obligations that follow, placing vetting, output review and preservation of privilege on the customer, so the firm is told what it keeps rather than simply warned off reliance. Clause 3.3 goes beyond a disclaimer. It limits attorney users to those licensed and in good standing in the United States and provides that use by paralegals must not constitute the practice of law.

That covers who may use the system and under what supervision, engages the unauthorized practice question directly, and carries a jurisdiction in the licensure requirement. No named authority is cited: no bar rule, ethics opinion or guidance document. That gap is recorded on the bar guidance signal.

No governance material is published. The security page and privacy policy were read in full, and the master services agreement clause by clause, along with the site navigation, footer and trust center entry point. There is no responsible AI page, framework or set of principles, no person or function named as accountable for model behavior, no account of what is tested before a model or feature change ships, and no certification such as ISO 42001.

The security page groups its content under software, governance and compliance, but every item there is an information security control rather than AI governance. Nothing addresses uneven output. On this product that has a specific form: a system analyzing spoken testimony will meet accents, speech patterns, interpreters and varying transcription quality, and nothing says whether performance across those conditions is measured or disclosed.

The controls are specific where it counts, and two elements are missing. All stored data is encrypted at rest using industry standard algorithms, and transport requires TLS with HSTS, which names the mechanism. Access is restricted to authorized users, with two factor authentication available on accounts. Assurance is ongoing: third party security experts run regular penetration tests, and the software supply chain is regularly scanned for vulnerabilities.

Deletion is contractual and immediate, with clause 6.4 requiring destruction of customer materials promptly after termination, subject to legal record keeping. On the other side, no incident or breach notification practice is published, and no subprocessor list exists for customer content; the agreement names third party large language model providers as a category without naming one. Retention during the term is not stated.

A complete allocation, published in full, with one provision that favors the customer. Clause 10.1 caps liability at twelve months of fees, waives consequential damages in both directions rather than only for the vendor, and carves indemnification obligations out of the cap, so the ceiling does not swallow the indemnity. Clause 8.1 is a mutual intellectual property infringement indemnity. Governing law is Washington, with arbitration before JAMS in Bellevue.

The gap is on the output side. Clause 9.6 disclaims all warranties on the accuracy, completeness or reliability of AI generated content and makes the customer solely responsible for it, so there is no warranty on output, and no insurance position is stated anywhere. A reseller channel operates on separate terms, which were not reviewed.

Integration is a named product line, but no system is named on any page read. The platform navigation lists a dedicated integration offering alongside the deposition, facts, agent and agencies lines, so connectivity is sold as a capability of its own. The separate offering for court reporting agencies implies a path for transcripts from the agencies into the firms. The integration page was not opened and may say more.

Nothing concrete was established. No document or case management system is named, nor any transcript or court reporting platform or repository, and no connector list, API or developer documentation is published. The product takes in transcripts produced by others and its output belongs in a case file, so how material arrives and where it goes afterwards is the integration question. Nothing published on the pages read answers it.

Nothing published addresses where the platform runs or how customers are separated within it. No cloud provider is named, no hosting region or country is stated, no residency option is offered, nothing distinguishes processing from storage, and no tenancy or isolation model is described. Cloud delivery is evident only by inference from a browser application at a login subdomain. The absence is consequential rather than formal on this product: the material processed is deposition testimony and case documents, which are among the most sensitive holdings a firm has, and a defense firm with client-imposed residency terms in its outside counsel guidelines has nothing published to assess.

A Vanta-hosted trust center exists and returned page metadata with no readable body, which is recorded as a retrieval limit rather than as an absence.

Two attestations are claimed with a trust center behind them, and the evidence itself is not published. The security page states SOC 2 Type II certification and describes what the standard demonstrates. It separately states that the company has audited its practices for HIPAA compliance, while noting that Litem is not typically a business associate under those regulations, which is a careful framing. Regular third party penetration testing is described as ongoing.

A trust center exists at the company's own subdomain and is signposted from the security page. What a reviewer would test is absent. No auditor or certifying body is named for the SOC 2 work, no report period, issue date or expiry is published, no scope statement identifies which services are covered, and no penetration test summary or date is given. The trust center is hosted on Vanta and returned page metadata with no body, so its access tier is not established and it may say more.

Outside model providers are acknowledged in the contract, one provider is named in a document that excludes the relevant data, and the two do not meet. Clause 5.3(f) of the master services agreement defines third party large language model providers as a term, so external models are in the path, but it names none. The privacy policy lists artificial intelligence and machine learning providers among the service providers that receive data, giving Google Cloud AI as the example.

That policy states at the outset that it does not apply to files, documents and other content uploaded for business customers, and routes those to the customer agreement. So the naming covers prospect and visitor personal data, not deposition transcripts and case files. A firm cannot tell from anything published which providers see client material. No model or version is named, nothing says where inference runs or what a provider may keep, and there is no commitment to give notice of changes.

No pricing information is published at any level. The navigation shows there is no pricing page to find: the platform menu lists the five solution lines plus security and contact, the resources menu lists careers, blog, help center and onboarding, and the footer repeats these with the privacy policy and agreement. Nothing names a plan, tier, unit of charge or structure. Nothing says whether the product is licensed per seat, per matter, per deposition or by transcript volume, which is the first question a buyer would ask of a product handling more than 10,000 depositions a month. Every call to action leads to a sales conversation through a scheduling link.

Two buyer segments and a lifecycle description, with the edges left open. The company serves law firms and court reporting agencies, and backs the second with its own product line. The agencies that produce transcripts and the firms that use them are different buyers with different needs. Coverage across a case is described in phases: starting in discovery, assisting through depositions in real time, and extending into later phases with agentic features, and the five solution lines map onto that.

Jurisdiction is stated indirectly but bindingly. Clause 3.3 requires attorney users to be licensed in the United States, which answers the question for a firm elsewhere. Named customers are large defense firms. The rest is missing: no practice area is named, no firm size is given, nothing distinguishes what plaintiff and defense sides get, and nothing says where coverage stops.