PatentWatch vs Tradespace: how they compare in 2026

The models are the entire product and the company says so directly. The seed announcement states that the manual work of reading claims, charting evidence, searching prior art and analysing potentially infringing products has historically been the bottleneck, and that large language models and modern search infrastructure are now at a point where that work can be automated. Every published module is a model function: interpreting a claim and decomposing it into elements, locating supporting passages in product documentation, scoring each element and the claim overall, running prior art and invalidity searches, and analysing a portfolio. There is no conventional product underneath and no licensed content asset, since the substrate is public patent data that only the models make usable for this purpose. Remove the models and nothing remains that a buyer would pay for, and the stated comparator is not other software but attorneys charting claims by hand. Verified 2 September 2026.

A conventional IP management system sits underneath a substantial and genuinely load-bearing AI layer. The substrate is a system of record: docketing, annuity tracking and maintenance cost modelling, portfolio composition reporting, workflow routing and eSignature-enabled assignments, all of which are what an IP management system has sold for decades and none of which needs a model. What the AI adds is real and is what the marketing leads with: disclosure packages generated from decks, specifications, flow charts and audio, invention reports summarising use cases, prior art and competitive advantage, patent drafting acquired with Paragon, a docketing assistant that summarises office correspondence and suggests responses, and licensing analysis that identifies likely licensees and builds claim charts. The company's own framing is that the AI is built into the architecture rather than added on, and a press announcement describes the product as a secure system of record combined with drafting technology, which is an accurate description of a platform where both halves are real. Verified 2 September 2026.

Grounding is demonstrated in an interface example and described nowhere. The home page carries a worked claim chart against a named patent number in which each claim element sits beside a quoted passage from product documentation, an analysis and a numeric score, with an overall match score for the claim, so a reader can see that outputs are meant to be traceable to a source passage. That is a demonstration rather than a disclosure: no published material explains what corpus the evidence is drawn from, how a passage is selected, what the scores measure, or how they are calibrated. No accuracy figure is stated. The home page carries three statistics under a heading about what is needed to monetise intellectual property, labelled faster claim charts, hours saved per chart and patent database size, and all three render as zero because they are animated counters that did not execute; the labels establish that figures are published, the values could not be read, and that is recorded as a limit on this reading rather than as an absence. The strongest accuracy claim located is in the seed announcement, which states the work can now be automated reliably and produce production-quality deliverables rather than rough drafts. That is an assertion without measurement. Verified 2 September 2026.

The mechanism is named and the question is answered directly rather than deflected. Asked on its own AI page how it prevents hallucinations, Tradespace states that it uses retrieval-augmented generation to limit responses to actual data and provides citations for responses, alongside guardrails intended to keep generated content grounded in fact. The Paragon acquisition announcement describes the drafting technology as offering transparent traceability to source materials so that teams can verify every claim, citation and technical detail. Naming the architecture rather than asserting accuracy is what puts this in the middle band. What is absent is measurement of any kind. No accuracy figure, benchmark, evaluation or test set appears anywhere, and the one number attached to output quality comes from a customer rather than the vendor, with the Director of IP at UT Austin quoted saying the platform gets his team 60 to 70 per cent of the way there, the equivalent of a first draft. That is a useful and candid figure and it is a testimonial rather than a measurement. The description of guardrails as state-of-the-art carries no content. Verified 2 September 2026.

Nothing published addresses oversight. Checked the home page in full, the news index, the seed announcement in full and the site navigation and footer on 2 September 2026, which together comprise the whole public site. No statement describes what the platform completes without a human, where a review point sits, what a user should verify before relying on a claim chart, or what the system does when the evidence for a claim element is weak. There is no terms of service and no acceptable use policy in which such an obligation could be placed. The gap is sharper than the grade alone conveys because of what the outputs are for: the company describes the product as generating claim charts used to identify litigation targets, support cease-and-desist correspondence and find counter-claims, and the seed announcement describes the outputs as production-quality deliverables rather than rough drafts. A platform producing assertions of infringement against named third parties, with no published account of who checks them, sits at the bottom of this axis. Verified 2 September 2026.

Review is asserted through the service layer and never described as a product control. The strongest oversight signal is commercial rather than technical: prosecution work runs through senior attorneys in a vetted network, and the platform page names attorney review and workflow approvals as stages. Inside the product, the customer can build no-code rules for disclosure routing and patent committee review, which is a configurable approval path a firm sets for itself. Against that, the docketing assistant is described as capturing office and counsel correspondence, summarising it, creating the resulting deadlines and suggesting responses, which is a system generating docket deadlines on a matter where a missed date is malpractice, and nothing published states whether a human confirms a created deadline before it enters the docket. No published material states what any module completes without review, where the checkpoint sits for a customer using the software without the attorney network, or what happens when the system is uncertain. There is no agreement of any kind published, so no review obligation exists in any document a buyer could rely on. Verified 2 September 2026.

No deployment evidence of any kind was located. Checked the home page in full, the news index, the seed announcement in full and the site navigation and footer on 2 September 2026. There is no customer page, no logo strip, no case study, no testimonial and no named user anywhere on the site, and the news section contains a single item, the April 2026 funding announcement. The three statistics the home page does carry are unattributed and rendered as zero because the counters did not execute, so even the vendor's own headline figures could not be read, and none of the three is a deployment measure in any event. The only outcome language located is a claim that a portfolio can start generating revenue within ten minutes, which appears in vendor-written copy on a third-party startup directory rather than on the company's own site. Nothing published names a customer, a sector, a portfolio size or a result. Verified 2 September 2026.

The reference base is broad, named and unusually well spread across sectors that are hard to win. Logos on the home page run to Georgia Tech, the University of Texas at Austin, the FDA, Enveda Biosciences, Unity, UNT Health and Raytheon, and the security page adds Northrop Grumman, the Department of Defense and the Department of Energy, stating that the DoD and DOE were founding customers. Two figures attach to named customers rather than floating free: UNC Charlotte is described as clearing a three-year technology backlog and posting 100 new case submissions in a few months, and Chun Kuo, Director of Intellectual Property at UT Austin, is quoted with both an outcome and a consequence, saying the platform gets his team 60 to 70 per cent of the way there and that without it he would need budget for more staff. Video case studies exist for Woodward and UNC Charlotte. A company announcement adds scale figures of more than 440,000 patents managed for over 80 organisations including 75 per cent of top US research universities. What holds this below the top band is that nothing is dated and no figure carries a method or measurement basis. Verified 2 September 2026.

Not one of the five limbs is addressed. Checked the home page in full, the news index, the seed announcement in full and the site navigation and footer on 2 September 2026. Nothing states whether customer content is used for training, nothing describes segregation between customers, neither privilege nor work product is mentioned, no retention or deletion position exists, and no model provider is identified. There is no terms of service, no privacy policy, no data processing addendum and no security page on the site in which any of it could appear, and a targeted search returned no such document either. The material a customer would upload makes this consequential rather than merely incomplete: the workflow requires a company's own patent portfolio and, for the analysis to be useful, its assessment of which competitors may be infringing, which is pre-litigation work product a patent owner would ordinarily treat as privileged and would not want visible to anyone else. Verified 2 September 2026.

One limb is met better than anywhere else in this corpus and two are absent entirely. The segregation architecture is described mechanically rather than claimed: a hub-and-spoke design provisions a separate instance of the model for each customer account, and the vendor states that no data is ever transferred from a spoke model to the hub or to another spoke. Tradespace also frames the confidentiality question in the terms that actually matter for patent work, addressing novelty rather than only secrecy, and stating that data uploaded does not constitute a public disclosure and is not discoverable by a third party. Training is addressed as an opt-in, with no training by default and only on explicit customer request. Model handling is disclosed, with Azure OpenAI Service named and Microsoft's data protection addendum linked. What is missing is fundamental. No retention or deletion position exists anywhere: nothing states how long disclosures, drafts or office correspondence are kept, or what happens to them when a customer leaves. Neither privilege nor work product is mentioned, which matters because the platform is sold to general counsel, chief legal officers and law firms as well as to R&D. And there is no published agreement, privacy policy or data processing addendum in which any of these commitments could be held. Verified 2 September 2026.

Nothing published addresses the advice line. Checked the home page in full, the news index, the seed announcement in full and the site navigation and footer on 2 September 2026. There is no statement that output is not legal advice, nothing on who may use the platform or under what supervision, no reference to verification duties, and no jurisdictional statement, and there is no terms of service or disclaimer page anywhere on the site in which such a statement could sit. The outputs make the absence material. A claim chart is a litigation instrument: it is the document a patent owner uses to allege that a specific product practises each element of a claim, and the company describes its charts as supporting cease-and-desist correspondence and the identification of litigation targets. The product also runs invalidity analysis intended, in the company's own framing, to establish that a patent will hold up in court. A platform generating infringement contentions and invalidity positions, publishing nothing at all about the professional obligations of the person who sends them, sits at the bottom of this axis. Verified 2 September 2026.

Nothing published addresses the professional responsibility questions, and this vendor raises more of them than most. Checked the home page, platform overview, Create, Manage, Enterprise AI and Security and Privacy pages, the counsel network page and the full site footer on 2 September 2026. There is no statement that output is not legal advice, nothing on who may use the platform or under what supervision, no reference to verification duties, and no jurisdictional statement. There is also no terms of service and no privacy policy anywhere on the site, so no document exists in which such a statement could sit. The gap is sharper here than on a pure software record because Tradespace markets itself as a full-service patent practice combining software with on-demand patent attorneys drawn from a vetted network of outside firms. That structure raises questions a buyer would expect answered somewhere: who the client is, whether an attorney-client relationship arises with Tradespace or only with the network firm, how conflicts are cleared across a network serving competing portfolios, and who bears the duty of candour on a filing the platform helped draft. None is addressed on any published surface. Verified 2 September 2026.

No governance material of any kind was located. Checked the home page in full, the news index, the seed announcement in full and the site navigation and footer on 2 September 2026. There is no responsible AI page, no governance framework or set of principles, no individual or function named as accountable for model behaviour, no account of what is evaluated before a model change ships, and no certification. Nothing anywhere addresses uneven output, which on this product would mean whether claim interpretation and evidence matching perform differently across technical fields, patent offices, document languages or the quality of available product documentation, all of which plainly vary and all of which shape which companies get named as infringers. The only statement in the vicinity is the seed announcement's assertion that the underlying technology has reached a point where the work can be automated reliably, which is a claim about the field rather than a description of any control the company applies. Verified 2 September 2026.

No governance material was located on any surface. Checked the home page, platform overview, Create, Manage, Enterprise AI and Security and Privacy pages and the full footer on 2 September 2026. There is no responsible AI page, no governance framework or set of principles, no individual or function named as accountable for model behaviour, no account of what is evaluated before a model or module change ships, and no certification such as ISO 42001. The nearest statement is a claim that the AI incorporates state-of-the-art guardrails to keep generated content grounded in fact, which names no guardrail and describes no testing. SOC 2 Type 2 maintained through Vanta is an information security attestation and is treated separately by this axis. Nothing anywhere addresses uneven output, which is a live question on a platform whose disclosure evaluation module scores inventions and whose licensing module ranks assets, since both produce judgements that shape which inventions get filed and which patents get monetised. Verified 2 September 2026.

None of the five elements is published. Checked the home page in full, the news index, the seed announcement in full and the site navigation and footer on 2 September 2026, and ran a targeted search for company security or legal documentation which returned nothing. There is no security page, no privacy policy and no data processing addendum. Nothing states any retention period or deletion practice, no subprocessor is named, no incident or breach notification practice is described, no encryption or access control is mentioned, and no trust centre exists. A login exists at the platform subdomain, so accounts and stored customer data plainly exist, and nothing published describes how either is protected. Under the rule that a site's own navigation establishes its page inventory, this is a thin site rather than an unreachable one: every page rendered cleanly and the whole public estate is a home page, a one-item news index and one article. Verified 2 September 2026.

Protection and isolation are described well and three of the five elements are absent. What is published is specific: encryption of all data at rest using AES-256, single sign-on through Okta, Microsoft and Google with others available on request, SOC 2 Type 2 compliance maintained through Vanta with a linked trust centre, hosting on Microsoft Azure with all servers stated to be in the US Northern Virginia region, and per-account dedicated model instances under the hub-and-spoke design. Against that, no retention period or deletion practice is stated anywhere for any category of data. No subprocessor list exists; Microsoft is identifiable as the cloud and model route and no other processor is named. And no incident or breach notification practice was located, with nothing stating whether or how quickly a customer would be told. The underlying reason for all three gaps is the same and is worth stating plainly: Tradespace publishes no terms of service, no privacy policy and no data processing addendum, so the documents in which retention, subprocessors and breach notification are normally found do not exist on the site. Verified 2 September 2026.

No agreement of any kind is published, so nothing addresses who bears the loss when the output is wrong. Checked the home page in full, the news index, the seed announcement in full and the site navigation and footer on 2 September 2026: the footer carries only a contact email address and a LinkedIn link, with no legal section, no terms of service, no privacy policy and no acceptable use policy, and a targeted search returned no such document from the company. There is no liability cap, no indemnity in either direction, no warranty, no service level and no insurance position on any reachable surface. The exposure this product creates is unusually concrete and unallocated: a claim chart asserting that a named company infringes is the basis on which a cease-and-desist letter is sent, and an error in it can expose the sender rather than the vendor. Nothing published says who carries that. Verified 2 September 2026.

Nothing published addresses who bears the loss when the output is wrong, because no agreement of any kind is published. Checked the home page, platform overview, Create, Manage, Enterprise AI, Security and Privacy and counsel network pages and the complete site footer on 2 September 2026: the footer carries Solutions, Our Platform, The Tech, Customers, Company and IP Resources, and there is no legal section, no terms of service, no privacy policy, no acceptable use policy and no data processing addendum. Two targeted searches returned nothing from the company either. No liability cap, indemnity, warranty, service level or insurance position exists on any reachable surface. The one commercial commitment published is a satisfaction promise rather than a liability term: on a first filing Tradespace states that if the customer is not happy it will not file and the materials are theirs regardless. For a platform holding unfiled invention disclosures for defence contractors, federal agencies and research universities, the absence of any published agreement is the single most consequential gap in this record. Verified 2 September 2026.

No integration into the systems legal or intellectual property work already lives in was located, and none is claimed. Checked the home page in full, the news index, the seed announcement in full and the site navigation and footer on 2 September 2026. No intellectual property management, docketing, document management or matter management system is named, no connector or integration is described, no API or developer documentation exists, and there is no integrations page in the navigation. The published workflow is self-contained: a user uploads patents and receives claim charts through the platform's own interface, reached by a login at the platform subdomain. Nothing describes export formats either, which matters because a claim chart's purpose is to be attached to correspondence or a filing rather than viewed in a dashboard. Verified 2 September 2026.

Integration is a genuine strength and it points in an unusual direction. Rather than connecting to the systems lawyers use, Tradespace connects to the systems inventors use, naming Slack, Gmail and Notion as sources it mines to surface ideas, and describing capture from pull requests, call recordings, schematics and design documents. That is the right architecture for a product whose bottleneck is disclosure capture. On the enterprise side the platform states support for integrations with CRM, CLM, ERP and other IP systems through APIs and connectors, with secure bidirectional exchange of data, which names the direction of flow rather than just asserting connectivity, and the Manage module describes integrations with annuity providers and AI-powered integrations with patent offices that capture maintenance events, deadlines and correspondence automatically. Single sign-on runs through Okta, Microsoft and Google. What is missing is documentation an implementer could use: no connector list, no named IP management or docketing system, no API reference and no developer surface was located. Verified 2 September 2026.

Nothing published addresses where the platform runs or how customers are separated within it. Checked the home page in full, the news index, the seed announcement in full and the site navigation and footer on 2 September 2026. 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 login at a platform subdomain, and even that is not stated as a deployment model. There is no security page or privacy policy in which any of it could appear. The company is described in third-party records with addresses in more than one country, and no first-party source states an operating location, so nothing published would let a buyer with data residency obligations assess the product at all. Verified 2 September 2026.

The tenancy answer is better than the residency answer, which is the reverse of most records here. Isolation is described concretely: each customer account receives a dedicated instance of the model under a hub-and-spoke design with no data moving between instances, and the vendor describes private instances as the mechanism that precludes public disclosure, which ties the architecture to the specific legal risk of patent work. Hosting is stated as Microsoft Azure and the location is given precisely, with all servers stated to be in the US Northern Virginia region. What is absent is choice. No alternative region is offered, no European or other option is mentioned, and no on-premises or customer-cloud deployment exists, so a customer with data residency obligations outside the United States has no published path. Nothing distinguishes processing location from storage location. The single-region answer is unusually specific and unusually inflexible at the same time, and for a platform serving research universities and multinationals filing in Europe and Asia the absence of any non-US option is worth a buyer's attention. Verified 2 September 2026.

No certification is claimed and no trust surface exists. Checked the home page in full, the news index, the seed announcement in full and the site navigation and footer on 2 September 2026, and ran a targeted search for company security documentation which returned nothing. Neither SOC 2 nor ISO 27001 nor any other standard is named anywhere, no auditor or attestation is referenced, there is no trust centre or portal, no badge or seal appears on any page, and no penetration testing or vulnerability programme is described. This is an absence rather than a retrieval limit: every page on the site rendered cleanly and the navigation establishes the full page inventory, so there is no gated or unreachable security surface behind which such material might sit. Verified 2 September 2026.

The attestation is real, named and reachable. Tradespace states that it partners with Vanta to maintain SOC 2 Type 2 compliance and links directly to a customer-specific trust centre, which is a route to controls rather than a badge on a page, and the security posture is described in specifics elsewhere with AES-256 at rest, named single sign-on providers and a stated server region. The claim is also contextualised by the customer base, with the company stating it works to these standards with clients including Northrop Grumman, Raytheon, the Department of Defense and the Department of Energy. What the top band asks for is not established. No auditor or certifying firm is named, since Vanta is the compliance automation vendor rather than the auditor. No report date, coverage period or scope is published, no trust services criteria are identified, and the access tier for the underlying report is not stated. Vanta-hosted trust centres render client-side and returned no readable content, which is recorded as a retrieval limit rather than an absence. No penetration testing programme is described. Verified 2 September 2026.

Nothing identifies what powers the product. Checked the home page in full, the news index, the seed announcement in full and the site navigation and footer on 2 September 2026. No model is named, no version, no provider entity, and nothing states whether any component is built in-house or reached through a third party. Nothing describes where inference runs, what any provider may retain, or whether customers would be told if the model set changed. The only reference to the underlying technology is in the seed announcement, which observes that large language models and modern search infrastructure have reached a point where this work can be automated; that is a statement about the state of the field rather than a disclosure of this company's stack, and it names nothing. There is no subprocessor list, no privacy policy and no terms of service in which such a disclosure could appear. Verified 2 September 2026.

The provider is named and the models are not. Asked directly which models it uses, Tradespace states a model-agnostic approach and says that by default it works with Microsoft Azure to access their OpenAI Services, linking to Microsoft's published data privacy terms for that service. That tells a reader whose infrastructure and whose model family sit underneath, and pointing at the applicable data protection terms rather than merely naming a vendor is more useful than most disclosures in this corpus. Where inference runs is stated through the hosting answer, on Azure with servers in the US Northern Virginia region, and the hub-and-spoke design tells a reader that the instance is dedicated rather than shared. What is absent is the specificity the top band requires. No model or version is identified, the model-agnostic claim is not accompanied by any list of what else can be plugged in, and no commitment exists to notify customers when the model or provider changes, which matters precisely because the architecture is designed to make swapping models easy. Verified 2 September 2026.

No pricing information is published at any level. Checked the home page in full, the news index, the seed announcement in full and the site navigation and footer on 2 September 2026. There is no pricing page, no tier or plan is named, no unit of charge is identified, and no structure is described, whether per user, per patent, per chart, per portfolio or by subscription. Nothing states whether onboarding is charged separately, and no trial or free tier is offered. The only commercial routes on the site are a Book a Demo control in the header and at the foot of the home page, and a contact email address. The one adjacent figure published concerns the cost of the problem rather than the price of the product, with the company noting that obtaining and maintaining a single patent can cost around fifty thousand dollars over its lifetime. No pricing row is written, because the rule that triggers one requires published structure and none exists. Verified 2 September 2026.

The charging model is stated and no number is attached to it. What a buyer learns without contacting anyone is the shape: prosecution is sold on a flat fee, with the vendor stating that the customer knows the cost before the work begins, and the whole commercial pitch is framed against the alternative of hourly outside counsel, with references to the law firm markup and to paying hundreds an hour for correspondence handling. There is a concrete free entry offer, with Tradespace preparing a full first application at no cost and stating that if the customer is not happy it will not file and the materials are theirs regardless. An ROI calculator is published for a buyer to model returns against their own portfolio. What is not published is any figure, tier, seat price, subscription rate or unit for the software itself, and nothing states how the platform licence relates to the flat prosecution fee or whether one can be bought without the other. The only route to a price is a demo request. Verified 2 September 2026.

Neither the buyer nor the boundary is stated. Checked the home page in full, the news index, the seed announcement in full and the site navigation and footer on 2 September 2026. There is no page addressed to any buyer segment: nothing distinguishes in-house intellectual property teams from law firms, licensing companies, universities or individual inventors, and no industry or technology field is named as supported. No jurisdiction or patent office is identified anywhere, so nothing states whether the claim charting and prior art work covers United States practice only or reaches the European Patent Office and other offices, which is the first question a portfolio holder filing internationally would ask. No portfolio size band is given and nothing states what the platform does not cover. What is published is a set of use cases rather than a coverage statement, running to licensing and litigation targeting, cease-and-desist support, counter-claim discovery, valuation in wind-downs and diligence, and most of that list appears in vendor-written copy on a third-party directory rather than on the company's own pages. Verified 2 September 2026.

Segment documentation is among the most thorough in the index. Six buyer roles each have their own page, covering IP leaders, tech transfer leaders, general counsels, chief legal officers, R&D leaders and corporate development, and five industries each have one, covering corporations, startups, universities, government and law firms. Four use cases are separately documented. That is a vendor that has thought carefully about who buys and has published the answer rather than leaving it to be inferred, and the customer roster bears it out across a federal agency, defence primes, research universities and venture-backed startups. Coverage of law firms as a buyer alongside in-house teams is notable given the product is also marketed as an alternative to outside counsel. What is missing is the boundary. No jurisdiction is stated anywhere, so nothing says whether the drafting and docketing modules handle European or international filings or only US practice, which the single US hosting region makes a live question. No portfolio size band is given and nothing states what the platform does not cover; trademarks and copyright are absent throughout without being excluded in terms. Verified 2 September 2026.

Checked the home page in full, the news index, the seed announcement in full and the site navigation and footer on 2 September 2026, and ran a targeted search for company legal documentation which returned nothing. Nothing addresses whether customer content is used to train or improve models, in either direction. No clause names training, model training, machine learning or model improvement in relation to customer content, and there is no terms of service, privacy policy or data processing addendum on the site in which such a term could sit. Nothing on any surface promises the opposite either. The silence has weight on this product because the material a customer uploads is its own patent portfolio together with, in use, its working view of which competitors may be infringing, and because vendor-written copy elsewhere describes a proprietary database of patent and product relationships that grows with each client, which is the shape of an accumulating dataset without any statement of what feeds it or what is done with it.

The structure is an opt-in rather than a prohibition, and the vendor is explicit about it. Asked on its own AI page whether it trains on customer or account data, Tradespace answers that by default it does not, and adds the quoted sentence, so training is available and switched on only at the customer's request. The security page states the same commitment in a slightly different frame, that data is never used for training without written consent, which adds a form requirement to the mechanism. A third statement on the AI page goes further than either and sits awkwardly beside them: asked whether using the platform constitutes a public disclosure, the vendor answers that customer data does not train any model, stated flatly and without the consent qualifier. Read together, the position is opt-in with written consent, and the unqualified sentence is best understood as describing the default rather than adding a prohibition. No agreement, privacy policy or data processing addendum is published anywhere on the site, so none of this is a contractual term and there is no document in which the consent mechanism is defined.

Checked the home page in full, the news index, the seed announcement in full and the site navigation and footer on 2 September 2026. Nothing addresses retention of any kind. No period is stated for uploaded patents, generated claim charts, prior art results or portfolio analyses, nothing describes what happens to that material when a customer stops using the platform, and no deletion route, export right or retention setting is published. The reason is structural rather than an omission from one document: there is no terms of service, no privacy policy and no data processing addendum anywhere on the site, so the documents in which a retention position normally lives do not exist, and a targeted search returned no such document either. A login exists at the platform subdomain, so uploaded portfolios and generated charts are plainly stored somewhere, and nothing published says for how long.

Checked the home page, platform overview, Create, Manage, Enterprise AI and Security and Privacy pages and the complete site footer on 2 September 2026. Nothing addresses retention of any kind. No period is stated for disclosures, generated drafts, prompts, office correspondence or portfolio records, nothing describes what happens to data when a customer stops using the platform, and no deletion route or export right is published. No retention setting is offered. The reason is structural rather than an oversight in one document: Tradespace publishes no terms of service, no privacy policy and no data processing addendum, so the documents in which a retention position normally lives do not exist on the site, and two targeted searches returned nothing from the company. What is published nearby concerns isolation rather than duration, with each account given a dedicated model instance and no data moving between instances, which addresses who can reach the data rather than how long it is kept.

Checked the home page in full, the news index, the seed announcement in full and the site navigation and footer on 2 September 2026. Nothing describes segregation between customers, isolation of one portfolio from another, or any access boundary. No tenancy model is stated and no permissions or role structure is described. The question has a specific edge on this product that nothing published reaches. The platform's value depends on mapping patents against competing products across a market, so two companies competing in the same technical field could plausibly both be customers, each having uploaded its portfolio and each being a potential target in the other's analysis. Nothing published states what separates them, and vendor-written copy elsewhere describes a proprietary database of patent and product relationships that expands with each client, which raises the question rather than answering it.

This is the first record in the pull to reach this value, and the mechanism is described rather than asserted. Asked how it ensures account data sent to AI models is not co-mingled, Tradespace describes a hub-and-spoke approach in which it provisions a separate instance of its hub model for each customer account, giving the quoted result, and states that no data is ever transferred from a spoke model to the hub model or to any other spoke model. The security page frames the same architecture against the risk that actually matters in patent work, stating that a private model per customer is what ensures uploaded data does not constitute a public disclosure. That is a documented separate-model design with the direction of data flow stated in both directions. Two limits belong on the record. The separation described is between customer accounts, and nothing addresses separation between matters or between clients inside a single account, which is the question a law firm using the platform would ask. And none of it appears in any agreement, since none is published.

Checked the home page in full, the news index, the seed announcement in full and the site navigation and footer on 2 September 2026, and ran a targeted search for company legal documentation which returned nothing. Nothing addresses disclosure to authorities or in response to legal process, and nothing addresses notice to the customer if a demand arrives. The question is not reached rather than answered adversely, and the reason is that no terms of service, privacy policy or confidentiality provision is published anywhere. The gap is worth naming on this product because the stored material is pre-litigation analysis: a customer's assessment of which third parties may be infringing its patents is exactly the material an opposing party would seek in discovery once enforcement begins, and nothing published states what the vendor would do if asked for it.

Checked the home page, platform overview, Create, Manage, Enterprise AI and Security and Privacy pages and the complete site footer on 2 September 2026, and ran two targeted searches for company legal documents. Nothing addresses disclosure to authorities or in response to legal process, and nothing addresses notice to the customer if a demand arrives. The question is not reached rather than answered adversely, and the reason is that no terms of service, privacy policy or data processing addendum is published at all. The nearest published statement concerns third parties in a different sense, with the vendor stating that customer data is not publicly disclosed or otherwise discoverable by a third party, which is a claim about the platform's architecture and confidentiality rather than about compelled disclosure. The gap is worth naming on this record because the customer base includes the Department of Defense, the Department of Energy, the FDA and defence contractors, where government demands for records are a foreseeable event rather than a hypothetical.

Checked the home page in full, the news index, the seed announcement in full and the site navigation and footer on 2 September 2026. The corpus is referred to and never identified. The home page carries a statistic labelled patent database with a figure in millions, but the counter renders as zero because it did not execute, so the scale claim exists and could not be read. No patent office, database, publisher or data supplier is named, no jurisdictions are listed, and no update cadence is stated. Two other bodies of material are plainly in use and are described nowhere: the product documentation and specifications from which evidence passages are drawn for claim charts, since the worked example on the home page quotes product literature directly, and the prior art sources behind the invalidity analysis. No licence, agreement or permission basis is stated for any of it.

The training corpus is described by category and no source or rights basis is given for any of it. Tradespace states that its AI tools were trained on knowledge gathered from millions of publications, patents, product specs, licenses, and search reports. Two of those categories are public records in most jurisdictions, and three are not: product specifications, licence agreements and search reports are ordinarily confidential or commercially licensed material, and nothing published states where they came from, on what basis they were obtained, or whether any of it originated with customers. That question sits directly against the vendor's separate statement that it does not train on customer or account data without explicit request, and a reader is left to reconcile the two without help. No database, publisher, patent office or data supplier is named, no jurisdictions are listed, no volume figure is given for any category, and no update cadence is stated. The platform separately consumes live patent office correspondence and prior art in operation, and nothing describes the provenance of that either.

Checked the home page in full, the news index, the seed announcement in full and the site navigation and footer on 2 September 2026. Nothing addresses whether material surfaced by the platform remains current or valid. The product does not retrieve primary law, so a citator has nothing to operate on and the limb does not bite in its usual form. The analogous question is live here and unaddressed: nothing states whether the platform checks that a patent it charts is still in force, whether maintenance fees have been paid, whether claims have been amended or cancelled in post-grant proceedings, or whether a competing product it identifies is still on the market. A claim chart built on a lapsed or amended patent is worse than no chart, and nothing published describes a status check of any kind.

Checked the home page, platform overview, Create, Manage, Enterprise AI and Security and Privacy pages on 2 September 2026. Nothing addresses whether material surfaced by the platform remains current or good. The question takes a particular form in patent work and is unaddressed in each of its forms: nothing states whether prior art returned in a disclosure evaluation is checked for currency, whether patent status and legal events are validated against office records when the Manage module reports on a portfolio, or whether the licensing module confirms that an asset it recommends for monetisation is in force and unencumbered. The docketing assistant does capture patent office communications and create deadlines from them, which is currency in the procedural sense, but nothing describes verification of what it captured. No treatment, status or validity signal is described anywhere.

Confidence is scored and surfaced, and nothing describes what happens at the low end. The worked claim chart on the home page shows a numeric score against each claim element, running from 95 down to 45 across the four elements displayed, alongside an overall match score of 85 out of 100 for the claim. Scoring each element separately rather than returning a single verdict is a meaningful design choice, because it shows a user which limb of a claim the evidence is weakest on, and the example deliberately includes a low-scoring element rather than a clean match. What is absent is everything around it. Nothing states what the scores measure or how they are calibrated, no threshold is published below which a match is not asserted, and nothing describes whether the system declines to chart a claim it cannot support or simply returns low numbers. There is no documentation surface on the site in which any of that could appear.

Checked the home page, platform overview, Create, Manage, Enterprise AI and Security and Privacy pages on 2 September 2026. Nothing describes what the platform does when it cannot ground an answer, and no confidence, coverage or grounding indicator is described as shown to the user. What is published is a constraint on generation rather than an account of behaviour at its limits: the vendor states that retrieval-augmented generation is used to limit responses to actual data and that citations are provided, and that guardrails keep generated content grounded in fact. One product behaviour comes closer than the rest and is recorded here, since the Create module is described as flagging missing information in an inventor's submission, which tells a user the input is incomplete rather than telling them the model is uncertain about an output.

Searched the AI Hallucination Cases database maintained by Damien Charlotin, and reporting drawing on it, on 2 September 2026 on the product names PatentWatch and Patent Watch. No court order, opinion or disciplinary record naming the product was located. This is a statement about the public record rather than a finding about the product, and two things bound it. The company is roughly a year old and its user base is not disclosed, so there has been little time and little visible surface for an incident to arise or be traced. And the failure mode this product would produce is not the one the tracker records: an unsupported infringement contention or a missed piece of prior art surfaces as a weak case, an invalidity ruling or a fee-shifting motion rather than as a sanctions order about fabricated citations, and no tracker indexes those.

Searched the AI Hallucination Cases database maintained by Damien Charlotin, and reporting drawing on it, on 2 September 2026 on the product and corporate name Tradespace and Tradespace, Inc., and on the acquired drafting brand Paragon. No court order, opinion or disciplinary record naming the product was located. This is a statement about the public record rather than a finding about the product. The tracker records fabricated citations in court filings, and this platform's outputs are patent applications, office action responses and docket entries filed at patent offices, so the forum that would surface an equivalent failure is the USPTO rather than a court, and its responses take the form of examiner objections, prosecution history consequences or an inequitable conduct allegation raised later in litigation rather than a published sanctions order. None of those is indexed anywhere comparable.

Checked the home page in full, the news index, the seed announcement in full and the site navigation and footer on 2 September 2026. No public material engages guidance from any professional body. Neither the USPTO Rules of Professional Conduct nor its guidance on artificial intelligence in submissions is named, no bar opinion is cited, and nothing addresses the obligations of a practitioner relying on a generated claim chart or invalidity position. Nothing addresses Rule 11 or the pre-suit investigation duty either, which is the provision that bites first on this product, since a claim chart is the usual evidence that a patent owner conducted a reasonable pre-filing inquiry before asserting infringement. There is no terms of service or disclaimer page in which any such reference could sit.

Checked the home page, platform overview, Create, Manage, Enterprise AI, Security and Privacy and counsel network pages and the complete site footer on 2 September 2026. No public material engages guidance from any professional body governing the platform's users. Neither the USPTO Rules of Professional Conduct nor its guidance on the use of artificial intelligence in filings is named, no state bar opinion is cited, and nothing addresses the duty of candour that attaches to a patent filing or the competence and supervision duties of a practitioner relying on generated drafts and machine-created docket deadlines. The absence carries more weight here than on a pure software record, because Tradespace markets a full-service patent practice delivered through a vetted network of outside firms and does not publish how professional obligations are allocated between itself, the network firm and the customer.

Time savings are claimed and the figures could not be read. The home page publishes three statistics under a heading about what is needed to monetise intellectual property, two of which are savings measures labelled faster claim charts and hours saved per chart. Both render as zero because they are animated counters that did not execute, so the claims are established and their values are not, and that is recorded as a limit on this reading. Nothing addresses how AI-assisted work is recorded, billed or disclosed to a client, and no per matter record of AI-assisted work was located. The question is live rather than academic on this product: claim charting is work that outside counsel bills for, often substantially, and the company positions itself explicitly against attorneys doing that work manually over weeks, so a firm using the platform faces exactly the compression this signal exists to track. Nothing published engages it.

Cost claims are central to the pitch and none of them is a disclosure record. Tradespace sells explicitly against the billable hour, describing itself as an alternative to the law firm markup, telling buyers to stop paying hundreds an hour for outside counsel to summarise office correspondence, offering flat-fee prosecution so that cost is known before work begins, and publishing an ROI calculator. The Manage module tracks outside counsel spend as a reporting feature. Nothing addresses how AI-assisted work is recorded or disclosed, and no per matter record of AI-assisted work was located. One structural note belongs on the record. This signal assumes a vendor selling to a firm that bills a client, and Tradespace mostly inverts that, selling to the in-house or university team that pays the bill, so its flat-fee model addresses the cost question at the commercial level rather than through disclosure. That inversion does not hold completely, because law firms are a named buyer segment, and for those customers the question the signal asks is live and unanswered.

Checked the home page in full, the news index, the seed announcement in full and the site navigation and footer on 2 September 2026, and ran a targeted search for company legal or security documentation which returned nothing. No subprocessor list exists in any form, no processor or vendor is named anywhere, and no model provider is identified. There is no data processing agreement, no privacy policy, no terms of service and no security page, so no forwardable client-facing artifact of any kind exists. There is no trust centre. A firm asked by a client which entities would process its patent portfolio and its infringement analysis could answer nothing from anything the company publishes, and would have to obtain every element of that answer by direct request.

The model route is named and nothing else is. Asked which models it uses, Tradespace states a model-agnostic approach and that by default it works with Microsoft Azure to access their OpenAI Services, and it links to Microsoft's published data privacy terms for that service rather than merely naming the vendor. A firm can therefore tell a client which provider processes its content and point at the applicable data protection terms, which is why the bottom value is not made out. It stops well short of the top value. No subprocessor list exists in any form, no other processor is named anywhere, and the model-agnostic claim means the named default may not describe what a given customer is actually running. There is no data processing addendum, no privacy policy and no terms of service, so no forwardable client-facing artifact exists at all; the SOC 2 Type 2 report behind the Vanta trust centre is the only obtainable document and its access tier is not stated.

Checked the home page in full, the news index, the seed announcement in full and the site navigation and footer on 2 September 2026. Nothing addresses disclosure of AI use to a court or patent office, and no exportable record of which model produced which determination, on what confidence, or who reviewed it is described. The product does generate the artifact that would need to carry such a record: a claim chart is filed, served or attached to correspondence, and the home page example shows each element with its supporting passage and score, so the traceability material exists inside the output. Nothing states that it is exportable, nothing describes a review or approval step, and nothing addresses what a party should say if asked how the chart was produced. The gap is pointed here because a claim chart is often the document offered to show that a reasonable pre-suit investigation took place.

Checked the home page, platform overview, Create, Manage, Enterprise AI and Security and Privacy pages on 2 September 2026. Nothing addresses disclosure of AI use to a patent office or tribunal, and no exportable per document record of which model produced which passage, what was retrieved and who reviewed it is described. The platform holds much of the raw material: citations are stated to accompany generated responses, the Paragon drafting technology is described as offering traceability to source materials, the docketing assistant captures all office and counsel correspondence for a case, and audit logs are named among the security controls. None is presented as a record of model use. The gap is worth naming because the duty of candour in patent prosecution attaches to what is submitted, and a platform that drafts applications, suggests office action responses and creates docket deadlines is generating exactly the material a later inequitable conduct challenge would probe.