A
AI Samurai

AI Samurai helps Japanese patent practitioners, corporate IP departments and inventors find prior art and prepare patent applications. AI Samurai Inc., based in Otemachi, Tokyo, became a Toyota Technical Development Corporation subsidiary in 2025 and still sells under its own name. AI Samurai ONE takes a written invention description and compares it with the company's own database, built from Japan Patent Office data, in tens of seconds.

It proposes International Patent Classification codes, builds a claim chart with matching passages for each element and rates the prospect of registration from A to D. Its detailed research tools cover Japanese, United States and Chinese patents.

ONE's drafting tool combines a base patent with up to 250 similar patents and places the pertinent passages in an editable specification that exports to Word. AI Samurai ZERO interviews inventors in a chat built on patent attorney expertise, shows prior art as the conversation runs and turns the answers into patent document text. A separately priced MCP server, ¥1,200,000 a year for 10 IDs before tax, links the patent database to Claude Code, Copilot, Codex or ChatGPT. The IP LANDSCAPE research service ended in July 2026.

Capability grades

All 15 axes, graded from public sources on the date shown. Hover a grade to see what the letter means on that axis.

AA on AI CentralityThe artificial intelligence is the product. Remove the models and there is nothing left to sell.

AI Centrality

How much of the product is actually AI. Whether the machine learning is the mechanism the buyer is paying for or a feature layered onto conventional software, and whether the vendor is specific about which is which.

ONE runs a similarity analysis of an invention description against every patent in its database and builds a claim chart that reports matching words or paragraphs for each element of a claim. The examination simulation proposes classifications and assigns one of four registration likelihood bands. Generative assistance turns a description and selected patent material into a draft specification. In ZERO, a conversational model asks the inventor questions, compares the answers with prior art and presents related publications while the description takes shape.

GPT is named in the ONE and ZERO procurement materials, and the company holds Japanese patent 6185209, granted in August 2017, for its intellectual property support system.

Source: Vendor Published
BB on Citation Accuracy and Hallucination DisclosureGrounding is real and documented, with linked primary sources and a described retrieval method, short of published accuracy figures an outsider can test.

Citation Accuracy and Hallucination Disclosure

Whether the vendor publishes measured accuracy on citations and assertions, grounds output to primary sources, and says plainly what its system does when it does not know. Legal has a documented public record of fabricated citations reaching filed briefs, so an untested claim of accuracy is not evidence.

Claim charts identify the document and passage matched to each element, and in ONE the landscape view opens the patent publications behind plotted points. Drafting draws on a base patent and a set of similar records the user picks. The terms commit the company to reasonable efforts to keep output accurate and disclaim any guarantee of accuracy, usefulness, legality or completeness. The MCP service gives an AI client direct access to patent records. No error rate for source matching accompanies the drafting workflow.

Source: Vendor Published
AA on Autonomy and Oversight ModelWhat the system runs alone, what constrains it, and how a lawyer checks it are all published: modes, thresholds, review surfaces, and the route a matter takes back to human judgment. A categorical limit on a named mode or tier, stating what its output may not be used for, meets the threshold limb without a number.

Autonomy and Oversight Model

What the system decides on its own, what a lawyer must approve, and whether the vendor documents where the review point sits. A tool that drafts under review and a tool that files without one are different products and different risks.

An inventor can begin with an idea and refine it in a chat, while a patent practitioner can work from a written description and control the patents used in a draft. ONE displays the cited passages and lets users change the resulting document in the application or in Word. An administrator can share a qualifying output with an external patent agent or attorney for a patent application under the terms. The product produces material for that workflow and leaves the decision to apply and the final submission with the applicant or representative.

Source: Vendor Published
BB on Operational and Outcome EvidenceReal deployment evidence with substance, short of full attribution or measurement: a named customer without figures, or figures without the named customer.

Operational and Outcome Evidence

Named, dated evidence that the product works in production at real firms or legal departments. Case studies with figures and identified customers count. Unattributed testimonials and launch announcements do not.

The Tokyo trial procurement catalog reports adoption by about 60 Japanese companies, including manufacturers, precision equipment makers and chemical businesses, and lists prior art investigations, infringement risk review and application preparation among the uses. The program listed ONE and ZERO in 2025, then ended their certification after a corporate restructuring changed eligibility. The vendor demonstrates how ONE breaks down a proposed invention and builds a claim chart.

The adoption figure is historical, and neither the listing nor the walkthrough gives a measured rate of wrong matches or of patent grants.

Source: Vendor Published
BB on Privilege and Confidentiality PostureSubstantive published commitments on confidentiality and training use, short of the full picture: commonly silence on segregation between users or matters, or on what the underlying model provider may retain.

Privilege and Confidentiality Posture

How client confidences are handled: attorney client privilege and work product treatment, segregation of one client matter from another, whether client data trains any model, and what the vendor commits to in writing rather than in marketing.

Article 10 treats invention content entered by the customer as confidential and limits its use to providing or using the service, for ten years after the contract ends. Article 17 lets the vendor back up use information for support or AI development and use it, without identifying any person or company, for statistics, marketing, service improvement and new services. No term keeps invention content out of AI development.

Administrators handle saving or deleting their own use information. Access at the level of a matter, and how each model provider treats unpublished invention data, are left open.

Source: Vendor Published
BB on UPL and Professional Responsibility PostureA real position is published on advice versus tooling, short of full treatment: commonly a disclaimer without the supervision and competence dimension, or silence on jurisdiction limits.

UPL and Professional Responsibility Posture

Whether the vendor is clear that it supplies a tool rather than legal advice, who its audience is, and how it addresses unauthorized practice of law, competence and supervision duties, and jurisdiction limits. ABA Formal Opinion 512 is the reference point. Where the advice line is not the duty a product raises, the axis is read through the nearest professional duty it does raise: judicial conduct rules and the reviewing duty for products sold only to courts, and the duty to bill for time actually spent for products that draft time entries.

The product supports preparation and first review of a patent application. It lets an inventor organize technical details and an attorney inspect the passages tied to each claim element. The terms permit sharing output with an outside patent attorney for an application and let a representative pass output to the applicant for that purpose, while disclaiming the accuracy and applicability of output. No mapping to the conduct rules for registered patent attorneys, and no automated review before filing, is offered.

Source: Vendor Published
CC on AI Governance and Bias DisclosureResponsible AI principles are published without a mechanism, a testing regime, or anything a buyer could audit.

AI Governance and Bias Disclosure

Published governance over model behavior: who owns it inside the vendor, what is tested before release, and what is disclosed about disparate output across matter types, parties, or populations.

ONE shows the text matches behind its examination simulation and separates related records in a visual landscape. ZERO shows users an example of the prompts and answers used to build a written invention description. Those controls give practitioners the material behind individual outputs. The company reports no accuracy testing broken down by patent office, language or technology class, and no process for reporting performance changes when its model providers change.

Source: Vendor Published
BB on AI Safety and Data StewardshipSubstantive published policy covering most of the ground, short of the full set: commonly no named subprocessor list or no stated incident practice.

AI Safety and Data Stewardship

Retention, deletion, access control, and what happens to prompts and documents after they are processed. Whether the vendor states its subprocessors and its incident practice, or leaves the buyer to assume.

Article 17 leaves saving, backup and deletion of use information to the customer's administrator. AI Samurai may back up that information for support or AI development, and after the contract ends it may delete everything on its servers at any time it chooses. Article 10 treats entered invention content as confidential for ten years after the contract ends. External AI services are listed separately with links to their terms, and the company may use use information in a form that identifies no person or company. No retention schedule for prompts and no deletion confirmation are offered.

Source: Vendor Published
BB on AI Liability and RecourseA real published position on liability, short of the full picture: commonly a stated indemnity without scope or caps.

AI Liability and Recourse

What the vendor stands behind contractually when its output is wrong. Indemnities, caps, carve outs, insurance, and whether any of it is published or only reachable through a negotiated agreement.

Article 20 of the terms caps AI Samurai's damages at the contract amount and lifts the cap for intent or gross negligence, with direct ordinary damages as the measure for specified breaches. The company does not guarantee that the application or its output is accurate, complete or fit for a purpose. Qualifying service interruptions can extend the subscription under a set calculation. The terms carry no indemnity for a wrong prior art match, a generated specification or a refused application.

Source: Vendor Published
BB on Practice Systems Integration DepthReal integrations exist and are documented, short of depth: named connections without a description of what they actually move.

Practice Systems Integration Depth

How deeply the product reaches into the systems legal work already lives in: document management such as iManage and NetDocuments, Word and Outlook, contract lifecycle management, matter management, e-billing, and court filing systems.

ONE exports a draft specification to Microsoft Word. The MCP server connects the patent database to model clients such as Claude Code, Copilot, Codex and ChatGPT, and AI Samurai shows a patent landscape dashboard built with Claude Code for competitive and portfolio analysis. Seminars, practical courses and company workshops support teams putting the connection to work. The connection brings patent information into the AI client a department already uses, and no connector to iManage, NetDocuments, docketing or billing systems is offered.

Source: Vendor Published
CC on Deployment Model and Data ResidencyCloud delivery is implied and neither the tenancy model nor the region is stated.

Deployment Model and Data Residency

Where the software runs and where the data sits. Multi tenant cloud, single tenant, private deployment, on premises, and whether region of residence is a published option or an enterprise conversation.

ONE and ZERO run as cloud software in the browser. AI Samurai operates from Tokyo and lists its external model platforms, Azure OpenAI, Claude through Amazon Bedrock and Google Gemini, in its terms. The MCP server connects to an AI client the customer chooses, which leaves part of the processing to the customer's own tools. Patent coverage follows the source documents available. Hosting and backup regions, model routing by region, a residency option per tenant and an installed version all go unnamed.

Source: Vendor Published
DD on Security Certifications and Trust CenterNo independent security attestation located.

Security Certifications and Trust Center

Independent attestation a buyer can pull without a sales call: SOC 2, ISO 27001, penetration test summaries, a trust center with current reports and named scope rather than a badge image.

AI Samurai commits in its terms to protective measures and disclaims complete prevention of unauthorized activity, and it keeps an information security policy beside its privacy notice. The privacy notice names who is responsible for personal information and how access and correction requests work. The external services list names the outside operators and links to their terms. No security certification or outside assurance report covers ONE or ZERO.

Source: Vendor Published
BB on Model Supply Chain DisclosureThe supply chain is partly disclosed: providers named without change notification, or architecture described without the providers.

Model Supply Chain Disclosure

Which models sit underneath, whose they are, where they run, and whether the vendor commits to telling customers when that changes. A legal buyer inherits every dependency it cannot see.

The external service list names Microsoft's Azure OpenAI, Claude through Amazon Bedrock and Google Gemini, and links to each provider's legal terms and privacy material. The ONE and ZERO catalog names GPT for assisted drafting without a model version. The MCP offering works differently, connecting patent records to an AI client the customer has already chosen. No mapping ties each application function to a model version or a processing region.

Source: Vendor Published
BB on Commercial TransparencyReal pricing is published for part of the range, with enterprise tiers withheld, or the unit and structure are stated without the figure.

Commercial Transparency

Whether a buyer can learn what this costs without entering a sales process: published rates, the unit being charged, what sits behind an enterprise tier, and what implementation adds.

Pricing for ONE and ZERO comes from the vendor's rate schedule or a written quotation. Tokyo's 2025 procurement catalog gave a reference figure from ¥3,900,000 including tax for 10 IDs, varying by seat count and options. The 2026 MCP release lists ¥1,200,000 a year for 10 IDs before tax and a standard initial fee of ¥150,000, with seminars, practical courses and company workshops sold beside the subscription. The terms let customers buy extra usage allowance for outside AI services if the included allowance runs out.

Source: Vendor Published
BB on Firm and Practice CoverageSegment and practice coverage is described with substance, short of the boundaries: what is supported is clear, what is not is left open.

Firm and Practice Coverage

Who the product is actually built for. AmLaw, midlaw, small firm and solo, in house departments, government and courts, and which practice areas are supported rather than merely claimed.

The products serve IP departments, in house research teams, patent firms and inventors in Japan. Customers come from automotive, machinery, electrical and precision equipment, chemicals, food and pharmaceuticals, construction, energy and IT, along with universities, government bodies, patent offices, law firms and research firms. ONE targets examination preparation, prior art analysis and specification drafting, and ZERO helps inventors, IP staff and new business teams turn an idea into an invention disclosure.

The detailed research tools cover Japanese, United States and Chinese patent documents. The MCP server adds PCT data in its launch materials and lists ten tasks, among them patent retrieval, competitive analysis, claim and specification drafting, office action responses, portfolio analysis and IP strategy.

Source: Vendor Published
Sources on file

3 public documents

The public pages on file for AI Samurai, with the recorded signals each one supports and the date it was last read. Open any of them and check the reading against the record.

Pricing

¥1,200,000/year for 10 MCP IDs, before tax

JPY, as published, never converted
  • ONE and ZERO are sold from the vendor's rate schedule or by quote.
  • A 2025 Tokyo catalog listed ONE and ZERO from ¥3,900,000 including tax for 10 IDs.
  • The separately sold MCP server is ¥1,200,000 per year for 10 IDs before tax.
  • The MCP server has a standard ¥150,000 setup fee before tax.

The ONE and ZERO catalog figure is historical and varied with the number of IDs and options. The 2026 MCP server rate covers 10 IDs for one year. The core terms let customers buy extra usage allowance for outside AI services if needed.

Implementation: The MCP server's standard initial fee is ¥150,000 before tax. An earlier trial promotion offered qualifying applicants a temporary reduction. AI Samurai shows no current setup fee for ONE and ZERO.

Note: The displayed annual figure is for the MCP server. ONE and ZERO pricing is set by a rate schedule or written quotation and can include an additional usage allowance.

Legal Signals

What each signal means

A signal records what public sources say on the date shown. It is not a grade and it is not a recommendation. Where a signal reads Not addressed, it means the index did not locate the material in public sources on that date, which is a statement about disclosure rather than about the product.

Confidentiality and Privilege

Client Data in Training

Can material a lawyer puts into this product be used to train a model?

Permitted, in the contract

The published agreement expressly reserves a right to train on customer content, with no opt out located. Any de identification, anonymization or aggregation qualifier is recorded in the summary.

Article 17 of the terms, last revised 31 October 2025, lets AI Samurai back up administrators' use information for purposes such as supporting the administrator or developing AI. Article 2 defines that information to include entered invention content. Article 17 also allows its use, in a form that identifies no person or company, for statistics, marketing, improving the service and developing new services, and no clause excludes model training.

Source: Vendor Published管理者のサポート又はAI開発等の目的As of Nov 20, 2026Evidence

Prompt and Output Retention

How long does the product keep what a lawyer typed, and can that be set to zero?

Disclosed without a period

Retention is acknowledged in public materials with no stated period.

Article 17 leaves saving and deletion to the administrator and lets AI Samurai delete everything on its servers at any time after termination. No fixed retention period applies to invention prompts, generated drafts or provider copies.

Source: Vendor Published任意の時点でAs of Nov 20, 2026Evidence

Ethical Walls and Matter Segregation

Does retrieval respect the firm’s ethical walls, or can the model read across them?

Not addressed

No located public material addresses walls or matter level segregation.

The terms distinguish administrators from users. Ethical walls, access by matter and permissions inherited from a firm's document system are not part of ONE or ZERO.

Source: Vendor PublishedAs of Nov 20, 2026Evidence

Third Party Request and Subpoena Notice

If someone subpoenas the vendor for a firm’s data, does the firm hear about it first?

Not addressed

No located term or policy addresses third party requests for customer data.

Article 10 bars disclosure of confidential invention information to third parties without consent, with exceptions for advisers and staff. The privacy notice lists exceptions for personal information. No term promises to tell the customer about a compulsory request.

Source: Vendor PublishedAs of Nov 20, 2026Evidence
Accuracy and Authority

Primary Law Corpus Provenance

Where does the law in this product come from, and does the vendor have the right to use it?

Sources named, basis unstated

Sources are identified without stating the license or rights basis.

AI Samurai builds its patent database from data obtained from the Japan Patent Office, and the detailed research tools cover Japanese, United States and Chinese patents; the MCP launch materials add PCT data. No reuse terms are given for each collection.

Source: Vendor PublishedAs of Nov 20, 2026Evidence

Good Law Verification

Does the product tell you when the authority it just cited has been overruled?

Not addressed

No located public material addresses whether authority is checked for subsequent history.

The examination simulation predicts the prospect of registration on an A to D scale and shows the publications and passages behind it. It does not verify a cited patent's legal status or follow how examiners or courts have treated it.

Source: Vendor PublishedAs of Nov 20, 2026Evidence

Refusal and Uncertainty Behavior

What does the product do when the answer is not in the corpus?

Confidence signal only

The product exposes a confidence or grounding score without an explicit abstention path.

The examination simulation puts each invention in one of four registration bands and backs it with matching passages in a claim chart. Neither the chat nor the drafting tool describes holding back an answer when matches are weak or a source is missing.

Source: Vendor PublishedAs of Nov 20, 2026Evidence

Fabricated Citation Record

Does a public court record exist addressing fabricated or hallucinated legal citations in output from this product?

None located

No court order, opinion or disciplinary record addressing fabricated or hallucinated legal citations produced by this product has been located as of the date shown. This is a statement about the public record on that one subject, not a finding about the product, and this signal is not a litigation history.

The AI Hallucination Cases database kept by Damien Charlotin has no entry naming AI Samurai ONE or ZERO. Both draft patent documents and prior art analyses for patent office work rather than court filings.

Source: Operator VerifiedAs of Nov 20, 2026
Professional Responsibility

Bar Guidance Alignment

Has the vendor engaged in public with the ethics opinions its buyers are bound by?

Not addressed

No located public material engages with bar or ethics guidance.

The terms let an administrator share output with an outside patent attorney for an application and let a representative pass output to the applicant. The AI workflow is not mapped to guidance from the Japan Patent Attorneys Association or any other professional body.

Source: Vendor PublishedAs of Nov 20, 2026Evidence

Billing and Fee Posture

Does the vendor address what happens to the bill when the work takes an hour instead of six?

Savings claims only

Public materials claim time savings without addressing billing or disclosure, and the product sits inside a fee relationship between a lawyer and a client where those savings would change the bill.

AI Samurai pitches time saved in patent investigation and drafting, with each lookup running across the whole database in tens of seconds. The workflow keeps no record of a patent attorney's AI assisted time or cost for disclosure to a client.

Source: Vendor PublishedAs of Nov 20, 2026Evidence

Outside Counsel Guideline Readiness

Can a firm get this vendor through a client’s AI clause without a bespoke negotiation?

Subprocessors listed

A current subprocessor or model provider list is published.

The external services register names Microsoft's Azure OpenAI, Claude through Amazon Bedrock and Google Gemini, with links to each provider's terms and privacy material. No outside counsel disclosure packet or controls for a particular client are offered.

Source: Vendor PublishedAs of Nov 20, 2026Evidence

Court Disclosure Support

If a judge’s standing order requires an AI disclosure, can the product produce one?

Partial record

Some elements of the record are available, short of a document level export.

ONE produces claim charts with matching passages and Word exports of draft specifications. Nothing combines AI activity, model details and each reviewer's actions into a single record for a disclosure.

Source: Vendor PublishedAs of Nov 20, 2026Evidence
Contact

Correct a record, or ask how something was graded

Every grade and every signal on this index is drawn from public sources and dated. If a record is wrong, out of date, or missing an artifact the index did not locate, send the source and it will be reviewed and the record redated. Vendors are welcome to submit documentation. Nothing on this index is for sale, including a listing, a placement, or a grade.

AI Legal Index

The AI Legal Index is an independent index that tracks changes to AI vendors in legal. It holds 617 vendors across 9 categories, each graded on the same 15 capability axes and recorded against 12 legal signals, from public sources. No vendor pays for inclusion, placement, or rating.

Index Status
Last index update
November 20, 2026
The AI Legal Index is an editorial reference. It is not a regulatory body, not a law firm, and nothing published here is legal advice or a recommendation to retain or avoid a vendor. Records are verified against published sources, bar guidance and public court records. Where a record reads not addressed, the material was not located in public sources on the date shown. See the Methodology page for evaluation standards and limitations.
© 2026 AI Legal Index
3801 N Capital of Texas Hwy, Ste E240 · Austin, TX 78746