DorothyAI vs IPRally: how they compare in 2026

The search models are the product. DorothyAI's own description is a suite built on artificial-intelligence search: Activ8 Novelty takes a plain-English description of an invention instead of keywords and Boolean operators, Activ8 Freedom reads the claims of granted patents for a product and its components, AskDot is a generative assistant over the patent database, and Adaptive Research Technology runs combinations of natural language processing, machine learning and legacy search models at once and surfaces results that more than one model agrees on. Take those away and what is left is a patent collection with no way into it that the vendor sells.

The machine learning is what the buyer pays for. IPRally's proprietary Graph AI, described as trained on millions of patent examiner citations, runs the search: a user enters a plain text description, a publication number, an image or a hand drawn search graph, and the model returns ranked prior art without Boolean construction. The company says this replaces query building rather than speeding it up, and the Agent page argues that generic large language models cannot search the patent index and that an agent is only as good as the search technology underneath. Without the models, what remains is a full text patent database with the Boolean search and filters IPRally offers as a secondary option, which is not what anyone buys.

The search method is described and nothing about it is measured. Adaptive Research Technology is set out clearly on the products page: several models optimized for different subject matter are queried simultaneously and results confirmed by more than one model are treated as the good ones, which the vendor contrasts with tuning a single model. Results are patent documents a reader can open. What is claimed and never tested is that this returns the best results every time and removes inconsistency across subject matter: no recall or precision figure, no benchmark, no test set and no error analysis is published. Nothing describes how AskDot's generated answers tie back to the patents behind them, which is the part of the suite where a wrong answer is hardest to see.

Grounding is documented and inspectable, and measurement stops short of a test an outsider could run. The Agent page says every search case can be opened, reasoning and citations are given per feature, and every claim assessment, feature mapping and relevance score traces to a source document in IPRally's patent index. The retrieval method is described: semantic search over graph representations, patent families, and following examiner citations outward from the strongest documents. One figure is published, a 14 percent improvement in search recall over a full text single search, but the baseline is IPRally's own, no test set is described, and no failure mode is named. The Generative AI Additional Terms of 14 March 2025 disclaim any warranty that outputs are correct and recommend checking them against a primary source. The site gives two corpus sizes, 120 million patents on the pricing page and 130 million on the Agent page. The product retrieves patent documents, not legal authority, so citator checks do not apply.

Human involvement is the pitch and the control structure is not described. The People+AI page puts more than 200 advisors alongside the tools and says the expert team runs the search and delivers the result, which describes a staffed service rather than a review step inside the platform. For the software a firm licenses, nothing published says what runs unattended, what a user must check before relying on a novelty or freedom-to-operate result, or what happens when the models disagree beyond the statement that agreement between models is how good results are picked. CrossCite is said to give control over references and intelligent filtering when building an Information Disclosure Statement, which is the nearest thing to a described review point.

The control structure is well described, with two gaps. What runs unattended is stated: Agent analyzes an uploaded disclosure, identifies claims and features, searches and builds the feature chart. Where it stops is stated too: Agent verifies its findings with the user before searching, and depth is a user setting, from quick triage to litigation grade digging. Every search case can be opened, reasoning and citations are given per feature, and output hands off into a full IPRally project for a professional to continue, described as designed for professional review, not blind trust. The Generative AI Additional Terms add that generative features are always marked before use, are enabled by an administrator and can be disabled at any time. Nothing published says what happens after the system is wrong, and no abstention path is described for Ask AI.

Three case studies are published and not one names anybody. The case studies page describes a leading technology firm that used an Activ8 Alignment report to design around a freedom-to-operate risk and then moved to quarterly monitoring, a prominent law firm that built an international filing strategy for an oil and gas client, and a mid-sized law firm that added patent landscapes and competitive intelligence to its practice. No customer is identified, no date is given, and no figure appears in any of them; the homepage testimonials carry photographs with no names or firms. What a buyer can take from the record is a description of how the reports are used rather than evidence of what they achieved.

One named customer has a figure, with no date or method behind it. The TLB case study names TLB GmbH, the central technology transfer office for universities in Baden-Wurttemberg, quotes Dr Frank Schlotter, Authorized Officer, and reports about a 50 percent cut in time spent on patent searches, linked to Smart search for query preparation and Ask AI during review. The PDF describes the earlier process of extensive keyword planning, manual Boolean queries and another AI tool the team was not satisfied with. Five more case studies name Unilever, RPX Corporation, Ossur, Metsa Group and Perl IP Consulting, and a testimonial on the Agent page names Josh Walling, Senior Patent Agent at Milwaukee Tool. No case study is dated, and the headline figure is the customer's own estimate of time saved. Only the TLB study was opened; the others may say more.

Nothing published says what happens to a search. This matters here: a freedom to operate query describes an unreleased product, and a novelty search describes an invention before it is filed, so the queries themselves are among the most sensitive things an IP practice holds. The Terms of Service, which govern the platform, address accounts, subscriptions, prohibited uses and intellectual property in the site's own content, and say nothing about the confidentiality of what a user enters or the results returned. The Privacy Policy covers personal information collected through the websites and does not reach search content. No statement on training, segregation between customers, privilege or work product is published.

The commitments a buyer most wants are in the agreement, and segregation is unanswered. The Terms of Service of 5 April 2024 say IPRally will not own user generated content, will not share, distribute or disclose it to any third party, and will use it only to perform the Service, and give the customer the right to delete all of it permanently. The Generative AI Additional Terms require any outside provider not to train on input or output or claim ownership, and say user generated content is never sent outside IPRally's cloud unless the customer approves. Privilege and work product are not mentioned anywhere read, and nothing addresses separation between users or matters within an account beyond company specific encryption keys and user level feature rights. That may suffice for an in house IP department, but IPRally also sells to law firms, and for them the matter level question is open.

The audience is left open at exactly the point where it matters. The products page addresses patent attorneys and their clients together. It frames Activ8 Novelty around the first question any inventor, scientist, entrepreneur or business person asks, whether an invention is patentable, which is a legal conclusion when someone who is not a lawyer acts on it. CrossCite is said to be tailored to meet the latest USPTO guidance on AI tools for Information Disclosure Statement submissions, the one place the site engages with a practitioner duty. No statement that the output is not legal advice, no description of the user's own review duty, and no jurisdiction limit is published.

What is published concerns the reliability of AI output, not professional responsibility. The Generative AI Additional Terms make users solely responsible for generated content, tell them to verify it independently, and place the entire risk on them, and the Agent page says Agent searches while the user decides. Nothing says IPRally is not a law firm or does not give legal advice, sets a jurisdiction limit, addresses a practitioner's supervision or competence duties, or engages with ABA Formal Opinion 512, USPTO practitioner guidance or any EPO or national equivalent. This matters because the Agent page markets a should we file signal to R&D and innovation teams described as needing no patent expertise, putting a patentability assessment in front of nonpractitioners with only a general verification disclaimer.

No governance position is published for a system whose results decide whether an invention is filed or a product is launched. Adaptive Research Technology is described as a design choice about how search models are combined, not as a governance regime: nobody inside the company is named as accountable for model behavior, nothing describes what is tested before a change ships, no evaluation is published, and nothing addresses whether results vary in quality by technology area, language or jurisdiction, which the vendor itself says is the weakness of single-model platforms.

Principles are published, with no mechanism behind them. The about page states three company values: transparency, defined as clear, explainable AI so a user always knows how decisions are made; simplicity; and reliability. The features page repeats explainability as a product property. That is a principle specific to AI. There is no responsible AI or AI governance page, no ISO 42001 or EU AI Act position, no one named as accountable for model behavior, no testing regime before release, and no evaluation of uneven output for a system that classifies technology and scores relevance. The marking of generative features and the customer's ability to switch them off are covered under oversight.

Nothing published states how long searches, uploads or reports are kept, who inside the vendor can reach them, or what is deleted when a subscription ends. The Privacy Policy is written for the websites: it lists cookies, log files and analytics in detail, and its headings for international transfers, access and correction and commitment to security carry no text at all on the page as served. The Terms of Service address billing, accounts and prohibited uses. No subprocessor list, encryption statement or incident practice is published.

Retention, deletion, access control, subprocessors and incident practice are all published and specific. The privacy policy, updated 12 December 2025, keeps personal data for at most twelve months plus a normal backup period after the reason for holding it ends. The Terms of Service let the customer permanently delete all user generated content, and the Data Processing Agreement requires deletion or return of personal data on termination. Access is limited to authorized staff bound by confidentiality, with SSO, optional adaptive multifactor authentication, user level feature rights and company specific encryption keys. DPA Annex B names nine subprocessors, Okta, Google Cloud EMEA, HubSpot, Intercom, Planhat, Mailgun, Zapier, Slack and Pendo, each with address, purpose, processing location and data categories. DPA clause 9 requires breach notice without undue delay and sets out four required contents, alongside a Vulnerability Disclosure Policy last modified 17 February 2026 and a public status page. Retention is framed around personal data, with no separate period for patent content or AI inputs, and the subprocessor list is dated January 2025.

The published position is a standard disclaimer and a cap set at what the customer paid. The Terms of Service provide the services as is and as available, disclaim all warranties including accuracy, reliability and fitness, and state that if liability is found it is limited to the amount paid for the products or services, with no consequential or punitive damages. No indemnity runs from the vendor, no warranty attaches to search results, and no insurance position appears. For a product whose output supports a filing or launch decision, nothing addresses what happens if a material reference is missed.

The allocation of loss is published and readable before signing, and runs entirely against the buyer. The Terms of Service cap liability at the amount paid in the twelve months before the event, and say IPRally is not responsible for damage caused by correct or incorrect data in the Service. The Generative AI Additional Terms provide outputs as is with no warranty of correctness, place the entire risk on the user, and disclaim liability for outside model services and for security failures at third party providers. The Data Processing Agreement has the same cap but excludes willful misconduct and gross negligence from it. No indemnity runs to the customer, no warranty covers output, and no insurance position is stated. A buyer can read the whole allocation before signing.

No integration into the systems a patent practice runs is published. Nothing names a docketing or IP management system, a document management system, an office add-in or an API, and CrossCite, which prepares Information Disclosure Statements, is described as generating the statements without saying how they reach a prosecution system or the USPTO. The platform is reached through its own web application.

The product largely stands alone. What is documented is export rather than integration: custom export to Excel, PDF and Word, sharing of searches, collections and monitoring within the account, optional single sign on, and an unspecified reference to tool interoperability on the features page. No connection to an IP management or docketing system, document management system or patent office filing system was found, and there is no integrations page. A company blog post dated 23 July 2026 says a new IPRally API has shipped and MCP support is coming; the future half is not evidence today. A help center collection on product news, data coverage and interoperability, with three articles, was not opened and could change this. Checked the home page, features, pricing, the product navigation, the legal hub and the help center collection list.

Nothing published states where the software runs or where a customer's searches are stored. The platform is delivered over the web at search.dorothyai.com; no tenancy model is described, no hosting provider or region is named, and the Privacy Policy's international transfers heading carries no text on the page as served. A practice with client-imposed residency requirements would have to establish all of it in a sales conversation.

Residency is answered plainly and tenancy is not. The privacy policy says IPRally's servers are in the European Union, and the subprocessor annex names Google Cloud EMEA Limited of Dublin as the main cloud provider processing in the EU, with a processing location for each of the other eight subprocessors. DPA clause 4 bars transfer of personal data outside the EU or EEA without the customer's prior written authorization and requires standard contractual clauses for any transfer. The AI question is answered too: the Generative AI Additional Terms separate In-Cloud Services, running inside IPRally's cloud, from External Services reached by API, say core models are hosted by IPRally, and commit that user generated content is not sent outside that environment without customer approval. Nothing says whether the platform is single tenant or multitenant, and the EU is the only location offered.

No independent security attestation is published. There is no security page and no trust center, no SOC 2, ISO 27001 or comparable standard is claimed anywhere, no auditor or penetration test is mentioned, and the Privacy Policy's own commitment to security heading carries no text on the page as served.

The certification is real and stated on the home and features pages as ISO 27001 certification of the information security management system. No scope, certificate date or period, or auditor is published on any readable page, so a buyer cannot confirm it is current. A trust center at trust.iprally.com, linked from the home page and footer and hosted on Vanta, returned an empty page frame on every attempt, and a search did not surface its contents. It may say more. On the pages that loaded, a named standard appears without scope, date or a route to the report.

The architecture is described and the models are not identified. Adaptive Research Technology is presented as the vendor's proprietary framework for querying combinations of natural language processing, machine learning and legacy search models optimized for particular subject matter, and AskDot is called a generative AI tool over the patent database. No model, version or provider is named, nothing states whether the generative layer runs on a third party's service or the vendor's own, and no commitment to notify customers when any of it changes is published.

The model supply chain is disclosed in a contract, not a marketing page. The Generative AI Additional Terms of 14 March 2025 name Anthropic PBC and OpenAI LLC as External Services reached by API, and Google LLC as an In-Cloud Service, covering Gemini models and Model Garden models that meet stated requirements, with links to each provider's terms. External Services host models outside IPRally's cloud and In-Cloud Services run within it, and the features page adds that IPRally hosts its core AI models itself. Every provider must meet minimum requirements: no training on input or output and no ownership claim over either. Providers may be added or removed at IPRally's sole discretion, but the list sits in a published agreement whose changes are posted, with material changes notified by email or at login under the Terms of Service, so customers can see when it changes.

The shape of the commercial offer is visible and the numbers are not. The Terms of Service state that parts of the service are sold on subscription, billed in advance on a monthly cycle and renewing automatically, that a free trial may be offered and converts to a paid subscription unless canceled, that fees are non-refundable, and that fee changes take effect at the end of a billing cycle. The products are separately named and split by function, the products page invites inquiries about Alignment Bundles, and the People+AI page offers an instant quote form for patentability, freedom-to-operate and invalidity searches. No rate, tier price or unit of charge appears anywhere, and every route ends at a demo booking or a quote request.

The unit and structure are published without a figure. Two plans are named, Individual for one user and Team for two to unlimited users, so the unit is the seat. A comparison table of about sixty items covers search, Boolean, review, AI assistants, monitoring, collections, collaboration, classification, coverage, security and support, each marked included, optional or coming soon. Paid extras are named: custom taxonomy AI classifiers, AI classification of monitoring results, simplified R&D licenses, SSO and adaptive multifactor authentication. A three day free trial is self serve. No rate, band, floor or currency appears, the Team plan routes to Contact Sales, and Agent is described as value based and agreed per customer. No volume bands or currencies are given, and nothing says what implementation adds.

The data behind the product is described specifically, and the limits are not. The published answer to what sources the search tools use names thirteen kinds: global patents, trademarks, SEC filings, licensing, agreements, assignments, litigation, financial data, market data by product category and supply chain information. A Jurisdictional Realities Database is described as weighing more than thirty factors per jurisdiction, covering innovation ecosystems, government policy and judicial systems, to inform filing strategy. The buyers addressed run from patent attorneys and law firms to in house teams, entrepreneurs and inventors. Coverage by jurisdiction or date range for the patent collection itself is not stated, nor how current it is or which technology areas the models handle less well.

Coverage is described in substance, and the outer boundary is left open. Five role pages address the patent searcher, analyst, manager, attorney and head of IP, and the Agent page explains what IP leaders, patent professionals, R&D and innovation teams, and law firms each get. Five use cases are supported: novelty and patentability, invalidity, state of the art, freedom to operate and classification. Coverage is 58 jurisdictions and 120 million or more patents with machine translation. Customers range from corporate IP departments such as Unilever, Bosch and Nestle to the Danish Patent Office, technology transfer offices such as TLB, and firms such as Perl IP Consulting and Laine IP. Some limits are stated per feature: Agent supports novelty and invalidity, with freedom to operate on the roadmap and Boolean monitoring coming soon. Nothing states limits on team or portfolio size, or that the platform covers patents rather than the wider IP estate.