Dodonai vs SmartDepo: how they compare in 2026

Every function a customer pays for is model output. Deposition summarization, medical chronology construction, OCR of scanned records including handwriting and tables, extract and draft agents, and semantic ediscovery that finds evidence by meaning rather than keyword make up the platform. Pricing charges by pages processed, with multipliers by process type, so the bill tracks inference directly. Without the models, what remains is a document store with a lookup box. Dodonai's FAQ calls it an AI document processing platform built for legal professionals.

The product is an AI deposition summary produced in minutes from a transcript, which the vendor says rests on patent pending technology. The newer platform layer is entirely model work: chat across every deposition in a case, a contradiction finder, theme extraction and automated memos. Human quality review sits on top of model output as a check, not as the production method. Without the models, what remains is the manual page and line summarizing the product exists to replace.

Deposition summaries carry page and line citations, so each point leads back to the transcript location it came from, and the transcript product adds automated Bluebook citations and an assistant that works over the full text. Chronologies and extracts carry page citations against the source document as well. Section 3(d) of the terms says that, given the probabilistic nature of machine learning, the services may produce incorrect output that does not reflect real people, places or facts. It tells customers to evaluate accuracy, including by human review. Dodonai claims its output is on par with a skilled human reviewer. No accuracy figure, test set or method is published.

Summaries carry hyperlinked page and line citations to the source testimony, the vendor publishes a comparison of its precise citations against competitors' broad topic ranges, and every summary is said to be reviewed by legal professionals before delivery. The marketing and the agreement say different things. The site advertises 100 percent guaranteed accurate page line citations and patent pending technology that can guarantee 100 percent accuracy. The governing terms at the same domain say the Services may contain errors, disclaim any warranty on the quality, accuracy, currency or completeness of the platform or its results, and make the customer solely responsible for checking work product before acting on it. No measured accuracy figure or test set is described. The failure modes the agreement names are audio quality, background noise, cross talk and speaker accent, which belong to transcription rather than summarizing.

Dodonai treats AI output as draft work product. Its Safety and Trust page says the customer's team verifies source citations, facts, dates and current legal authority before any use or release. Automated source links, the page adds, do not replace citator checks or the firm's own research and supervision. Page and line citations give the reviewer the path back to the source, and every action is logged. The custom agents sold through AI Managed Services go further. They draft and never send, attorneys authorize every action that leaves the firm, and an agent that is not confident asks instead of guessing. Inside the App, no threshold at which processing stops for review and no route after a wrong summary is described.

SmartDepo says legal professionals review every summary before it reaches the customer and lists humans in the loop as one of three trust pillars, a gate on every output. Section 6 of the governing terms adds a second check by making the customer solely responsible for verifying accuracy and completeness before acting. There is no threshold at which the system defers, no description of what reviewers check or against what standard, no stated reviewer qualification, and nothing on what happens when an error reaches a filed brief.

One quotation is published, from a commercial litigator identified only as Ben, saying that on the first day Dodonai summarized every deposition transcript in a case in seconds, saving hours of work and client costs in the hundreds. No organization is named, and there is no logo strip, case study or customer count. The litigation cost page claims a 300 page deposition summary that costs $1,500 to $7,500 from a service vendor costs under $10 in Dodonai, with no customer or method behind the figure. Comparison pages for deposition and chronology software are Dodonai's own category content.

Eleven customer firms appear as logos, including Mueller Law, Mann and Potter, Explico, Windsor Troy Law, the Hernandez Legal Group, Rencher Law Group, Amini and Conant, Struble Cohen Trial Lawyers, Broussard Knoll and Conforto Law Group. Four testimonials carry names tied to firms: Wolfgang Mueller of Mueller Law, Delaney J. Miller of Windsor Troy Law, Matthew Struble of Struble Cohen, board certified in civil trial and appeals, and a reviewer at Conant. The quotes include operational detail, such as about 30 minutes to process a transcript and only minimal changes needed on review. No time or cost saving is tied to a named firm, and nothing is dated. The vendor also publishes a customers page.

Section 3(e) of the terms says both parties acknowledge that content may include confidential and privileged information and that Dodonai will use all reasonable efforts to keep it confidential. It goes on to state that both input and output are attorney work product and that Dodonai shares a common interest with its users. Workspaces are isolated for each client, case data is siloed, access is role based and every action is logged. Under section 3(c), Dodonai works only with LLM partners that disclaim using the content to improve their own services and delete customer data after a reasonable time, 30 days for OpenAI. Dodonai may use customer data to improve its own services only where the customer opts in. No retention period applies to Dodonai's own storage, and OpenAI is the only partner named.

Section 9 of the governing agreement is a conventional mutual confidentiality clause: customer content is the customer's confidential information, the recipient must use reasonable care, and disclosure is limited to staff and subcontractors under equal obligations. Privilege and work product are never mentioned, on a product whose entire input is deposition testimony. Nothing addresses separation between customers or matters. No retention period is stated; section 2.2 instead reserves to the vendor the right to set how long it keeps customer content. Section 4.2 permits training rather than barring it. Section 2.3 bars submitting protected health information or sensitive personal data unless an Order expressly allows it, a live constraint for medical malpractice and personal injury depositions.

Dodonai's Safety and Trust page calls AI output draft work product and puts verification of citations, facts, dates and current authority on the customer's team before use or release. The same page walks through ABA Formal Opinion 512, mapping the duties of competence, supervision and confidentiality onto AI use, and offers the agent audit trail as support for supervision. The terms do not say the services are not legal advice, and they name no jurisdiction limit. Section 3(e) calls input and output attorney work product and asserts a common interest with users. The audience reaches beyond lawyers, with material for paralegals, court reporters, IME companies and IME doctors, and section 1 sets the minimum user age at 13, with parental permission under 18.

The governing terms of service contain no statement that output is not legal advice, no competence or supervision language, no reference to any rule of professional conduct, and no jurisdiction limit. Section 6 requires the customer to verify accuracy before acting, framed as an accuracy precaution rather than a professional one. The marketing says the product was built by attorneys for attorneys and founded by a practicing civil rights attorney. Its output is meant to be copied into motions and briefs and submitted to courts. The site's claim that legal professionals review every summary is a quality control statement, not a position on advice.

Section 3(d) of the terms acknowledges the probabilistic nature of machine learning and the chance of incorrect output, and section 3(b) explains that output may not be unique across users. The Safety and Trust page sets out a safety layer for the custom agents. Citations are verified against Westlaw, Lexis or the primary source, the opinion text sits beside each summary, attorneys authorize every outbound action, and an audit trail records what each agent saw, produced and changed. No one inside Dodonai is named as accountable for model behavior, no evaluation before release is described for the App, and no testing results are published. Nothing addresses bias or uneven output across injury types, record types or patient populations.

Output is copied into court filings, and no responsible AI statement, governance framework, named owner, description of testing before release, evaluation result or bias statement is published. The nearest published material is section 4.2 of the agreement, which states that Rev's speech recognition and other AI models are proprietary, maintained locally and not shared with any third party, a data handling statement rather than a governance regime.

Data is encrypted with AES 256 at rest and TLS 1.2 or higher in transit, from the moment documents leave the browser. Workspaces are isolated for each client, access is role based and every action is logged. Section 5(a) of the data processing addendum lists the required security measures across personnel, facilities, hardware and software, storage and networks, access control, monitoring and logging, breach detection, incident response and encryption. Section 5(c) requires notice of any personal data breach by Dodonai, its LLM partners or anyone acting for it without undue delay, and section 8 returns or deletes customer data on termination or reasonable request. No retention period for Dodonai's own storage during the term is published, and no subprocessor list names more than OpenAI.

Section 2.2 of the governing agreement reserves to the vendor the right to establish general practices and limits including the maximum period it will retain customer content, and no period is published. No deletion commitment is stated. Security is addressed only in section 9.3, which promises measures in accordance with industry standards and applicable law, with no controls named. No incident or breach notification practice is published. Subcontractors are acknowledged in section 1.3, including individual freelancers who perform human-based services, with none named. A data processing addendum and a business associate agreement are incorporated by reference and published on the parent's domain. Section 2.3 places the backup obligation on the customer and excludes vendor liability for loss of data.

Section 7(b) of the terms provides the services as is and disclaims every warranty, including that the services will be accurate or error free or that content will stay secure and unaltered. Section 7(a) runs one way: the customer indemnifies Dodonai, and no indemnity is given to the customer, intellectual property claims included. Section 7(c) caps aggregate liability at the greater of twelve months of fees or $100 and excludes indirect and consequential damages, loss of data included. No insurance is stated, and the customer carries the loss when output is wrong.

Section 10 commits the vendor to keep insurance at stated minimums for the term: commercial general liability of one million dollars per occurrence and professional errors and omissions cover for the Service, including network security and data protection liability, of at least two million dollars per claim and four million in aggregate. The indemnity in section 7 runs only from customer to vendor, with nothing to the customer, including for third party infringement. Section 8 caps liability at fees paid in the previous twelve months for the services behind the claim. Section 6 disclaims all warranties, including accuracy. It carves out a warranty said to be at section 6.2, but no such subsection appears in the published document, so the one promised warranty cannot be read. No insurance certificate or carrier is named.

Uploaded files are the working material, and Dodonai's case management and document management pages present it as a processing layer beside the systems a firm already runs rather than a connected one. No document management, case management, email, word processing or signature product is named, and there is no integrations page or developer documentation. Section 2(c)(iv) of the terms refers to extracting data through the API as permitted, so an API exists, and the enterprise tier lists custom integrations without detail.

No document management system, case management platform, transcript repository, word processor or electronic signature product is named on the home page, the pricing page or in the terms, and there is no integrations page or developer documentation in the navigation or footer. Section 1.2 of the agreement mentions APIs made available to use the services without saying what they connect to. The workflow described is manual: upload a transcript, receive documents, copy and paste citations into briefs. The Products menu also lists a Court Reporting Solutions page.

Each client gets an isolated workspace, with case data siloed, role based access limiting what team members see, and every action logged. That separates customers from each other. No region, country, hosting provider or processing location is published for the App, and the terms say only that the services may be used in geographies Dodonai currently supports. Custom deployments from AI Managed Services can run in the customer's own cloud account or Dodonai's, with hosting location, model processors and retention set in a blueprint for each build.

Nothing is published about where the software runs or where transcripts sit. No region, country, hosting provider or data center is named on any surface, and no tenancy model is described: nothing states whether the platform is single or multi-tenant or how one customer's transcripts are separated from another's. The agreement is silent on both, referring only to Rev's servers in the context of storage allocation limits at section 2.2. There are no deployment options, no private or dedicated tier, and no residency commitment of any kind.

Dodonai states SOC 2 coverage and HIPAA compliance in its hero, its security section and its platform description, with a business associate agreement on request. The Safety and Trust page invites customers to request report information confirming the covered system, reporting period and access conditions. Section 1(i) of the data processing addendum commits Dodonai to cooperate with audits where the law requires, or else to share a summary of third party audit results. The claim does not say Type I or Type II. No auditor, scope, observation period or report date is published, there is no trust portal, and penetration testing is not mentioned.

No independent security attestation is published. No SOC 2, ISO, HIPAA certification or any other framework is claimed anywhere on the site, there is no trust center or security page, no auditor is named, no report is offered at any access tier, and the site carries no security badges. The agreement offers only section 9.3, promising security measures in accordance with industry standards and applicable law without naming a standard. A published customer testimonial refers to extensive discussion with the team on security before purchase, so security assurance comes through a sales conversation rather than a published document.

Section 2(c) of the terms requires customers to follow the terms and policies of any LLM partner used for processing, OpenAI included, and links OpenAI's policy pages. Section 3(c) limits Dodonai to LLM partners that disclaim using content to improve their own services and delete customer data after a reasonable time, 30 days for OpenAI. Section 1(g) of the data processing addendum commits to notice of any additional LLM partner and gives fifteen days to object on reasonable data protection grounds. Dodonai can then decline the partner or offer an alternative, among four cure options, and the customer may terminate with a refund of prepaid fees if the objection is not resolved within thirty days. No model is named, OpenAI is the only partner identified, and where inference runs is not stated.

The home page carries an icon labeled OpenAI Zero Data Retention among its three trust pillars. It names a third party model provider and the retention terms agreed with it, though no text explains the scope or which features it covers. Section 4.2 of the agreement describes the vendor's own speech recognition and generative models, stated to be proprietary, kept locally and not shared with any third party. So both a named external provider and proprietary internal models are in use. No specific model or version is named, no location is given, no subprocessor or model provider list is published, and no commitment to notify customers of model or provider changes is published.

Four tiers are published. Billed annually, $300 buys 4,800 credits a year ($25 a month), $600 buys 24,000 ($50 a month) and $1,000 buys 120,000 ($83 a month), with a custom enterprise tier above them. A page is defined as about 400 tokens of text, and a table sets the cost per page at each volume by process type, with standard summaries running from 6.3 cents down to 0.8 cents. Custom summaries use credits at twice the standard rate and advanced analysis at three times, while OCR and transcription cost nothing on any tier. All nine capabilities come with every plan, without seat fees or feature gates, and enterprise adds custom integrations, service level guarantees and dedicated onboarding. A free trial of seven days and 100 pages opens from a sign up form, and section 4(c) of the terms promises fourteen days' notice before a price rise.

Two purchase routes are published side by side. The platform subscription is 99 dollars per user per month and reduces summaries to a flat 25 dollars each at any volume, with the included capabilities itemized as AI deposition chat, automated memos, contradiction finder, and theme extraction and clips. Standalone summaries are published as a volume table with five bands: 85 dollars for one to nine, 75 for ten to nineteen, 65 for twenty to thirty, 55 for thirty-one to forty, and 50 for forty-one and above. An interactive estimator on the page computes a monthly total against the pay-per-summary alternative and shows the saving. A seven-day free trial is offered with no credit card required and cancellation at any time, and signup is self-serve, so no sales contact is needed. The unit is one summary per deposition transcript. No figure for implementation is published, since none is required.

Six practice areas each have their own page: personal injury, medical malpractice, mass tort, product liability, workers' compensation and disability benefits. Eight use cases have pages too, including medical record review, IME reports, demand letters, expert witness preparation, contract review extraction and litigation cost management. Six industries are addressed separately, reaching past law firms to court reporters, IME companies, IME doctors and medical record retrieval businesses. The practice areas sit almost entirely on the plaintiff side. No firm size or segment is stated, in house and government use are not addressed, and nothing says which matter or document types the product does not handle.

The buyer is described as attorneys and litigators generally, with the marketing framed as built by attorneys for attorneys, and a Court Reporting Solutions page addresses court reporting agencies as a second segment. No practice area is named anywhere as supported: the named customer firms span trial and personal injury practices but the vendor itself claims no area of specialization, no matter type and no transcript type. Firm size is not addressed, in-house and government use are not mentioned, and nothing states where the product stops. The product covers deposition transcripts only, but that scope is implied by the offering rather than stated as a limit.