M
Minesoft Origin

Minesoft Origin is a patent research platform for patent attorneys, corporate IP teams and researchers, from Minesoft Ltd of Richmond upon Thames, which launched it in 2023 as a product separate from PatBase. A user describes an invention in plain language, and a vector model Minesoft trained in house finds similar inventions among more than 190 million publications from over 110 authorities, beside conventional Boolean tools.

Origin highlights relevant passages, writes plain language summaries, answers questions about a patent in Doc Chat, compares claims, sorts results into the team's own taxonomy and sends AI alerts with relevance ratings.

The MAIA add on runs novelty, invalidity, freedom to operate and state of the art workflows, with human checkpoints at the invention summary, the research plan, the shortlist and the report, and links every finding to its source. It runs on Minesoft's own servers in two UK data centers, while Doc Chat and summaries use outside models under zero retention terms.

Switzerland's Federal Institute of Intellectual Property lists Origin among the tools it uses for commissioned patent investigations with the client's consent. Origin is sold on annual quotes after a demo or trial, with MAIA as an extra entitlement.

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.

A vector model Minesoft trained in house matches a plain language invention description to passages and related patent records. Users can refine results, request document summaries and compare claims across candidate publications. The product can also learn a private classification from the user's choices and tags, and AI alerts learn what matters to a user and rate the relevance of new patents. The optional MAIA assistant structures an invention, proposes a research plan, retrieves and reranks materials, and produces technical findings linked to source publications. Conventional Boolean retrieval is available beside the AI workflows.

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.

Document Chat operates within the current patent text and metadata. An analyst can inspect relevant passages in the original record and compare overlapping or distinctive claim features across publications. The optional MAIA workflow ties findings to source documents and uses curated patent data for status and family references. The platform does not supply an independent measurement of citation accuracy. The annual agreement limits liability for access to and quality of patent office data.

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.

The user can revise the invention description, choose the documents reviewed, and inspect the relevant passages before using findings. MAIA pauses at checkpoints for the user to review the features and planned research, and gives the practitioner an opportunity to inspect the shortlist. The workflow then generates a report for professional review. Minesoft leaves material legal conclusions to patent counsel. Monitoring or classification can be automated without giving the software authority over an application or legal opinion.

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 Swiss Federal Institute of Intellectual Property names Origin in its consent form for commissioned patent work, giving clients a choice about AI tools when sensitive invention information is involved. Minesoft's MAIA walkthrough shows a novelty case that screened 92,112 documents, reranked 35,638 by meaning and built a shortlist and report. Minesoft reports more than 90,000 active users across its products and answers 50 to 80 security assessments a year.

Its time saving figures are estimates and do not measure missed prior art or attorney agreement across fields or patent offices.

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.

Mutual confidentiality in the annual terms is subject to defined exceptions. Under the AI security policy, shared models do not train on client data, and user requests and results stay inside Minesoft. The optional private classifier learns from the customer's tags and activity inside its own configuration. MAIA processes invention disclosures on servers Minesoft runs in the United Kingdom, and Minesoft points unpublished disclosures to MAIA rather than Doc Chat.

Doc Chat sends questions anonymously to an outside model under zero retention terms. Senior support staff need the user's written consent to view saved requests.

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.

AI tools support patentability triage, comparison of claims, and review of patent status and timelines. The company explains that its findings assist practitioners and should be reviewed by patent counsel for legally material decisions. MAIA includes human checkpoints. The platform displays patent office information and links to the source, leaving interpretation and filing with the user. The product does not map these steps to a named professional rule for patent agents or attorneys.

Source: Vendor Published
BB on AI Governance and Bias DisclosureA published governance framework with real substance, short of testing results or a named owner.

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.

Taxonomies and saved results give users a consistent way to organize a project. The MAIA workflow explains why references were shortlisted, lets a user review features and planning, and links outputs to the underlying patent records. The intended controls apply at different stages of patent analysis. The available materials do not give a bias evaluation by patent office, technology area, or language, or a schedule for disclosing changes to the deployed model.

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.

Minesoft says query history is stored in its own environment and users can delete it at any time. Its security materials separate the shared retrieval models, the private classification model, MAIA processing and the outside services behind Doc Chat and summaries, which run under zero retention terms. Only a few senior staff can reach customer data for support, and only with the user's prior written consent. How long MAIA prompts and reports stay in Minesoft's environment is not stated, and a customer's private classifier and saved material remain available for later use.

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.

The annual subscription terms warrant that the product operates materially according to its description or technical specification. They limit liability by reference to the purchased license fee and include a specified intellectual property indemnity. Availability and quality of third-party patent office data are treated separately. The company advises attorney verification for material legal findings. The general terms do not supply a separate guarantee or remedy for an incorrect AI novelty or invalidity result.

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.

Origin lets practitioners save and share results, export reports and set alerts and monitors from a project, and MAIA findings feed back into Origin in three report formats, an executive summary, a standard report and a full dossier. Minesoft also sells its patent data through Minesoft Connect, an MCP service for AI agents, and a patent data API, separate from the Origin interface. MAIA adds role based access, single sign on and a full audit trail. No native connector to a law firm's document management or docketing system comes with the core subscription.

Source: Vendor Published
BB on Deployment Model and Data ResidencyDeployment model is stated clearly with partial residency detail, or residency is offered without the processing location being addressed, or the tenancy model is stated on its own with no residency detail published.

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.

The patent platform runs on Minesoft's private cloud in the United Kingdom under UK GDPR, with AI retrieval, classification and summaries on its own servers. MAIA runs on servers Minesoft controls in two UK data centers, a primary and a secondary, and its processing stays in the UK even after failover. Doc Chat and summaries send individual requests to outside AI providers, in UK or EU data centers where the provider allows it, and those providers may reroute traffic to the United States or Asia when capacity runs short. Residency therefore differs between MAIA and the other AI features.

Source: Vendor Published
BB on Security Certifications and Trust CenterCertification is real and stated, short of accessible evidence: a named standard without scope, date, or a way to obtain the report.

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.

The data center hosting Minesoft's applications operates under ISO 27001 certification. Minesoft's own information security management system completed its stage two audit and was recommended for certification, with the certificate expected in early November 2026, and no issued certificate is shown yet. Minesoft describes least privilege for personnel, offers further certification documents on demand and answers security questionnaires in writing. The hosting provider's certification scope is distinct from the Origin service as a whole.

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.

Semantic retrieval runs on a vector model Minesoft trained in house in advance on public patent data, separately from user requests. MAIA runs on Minesoft's own infrastructure without third party models, and its model goes unnamed. Doc Chat and summaries use outside providers, also unnamed, through APIs under zero retention terms. Classification models train on patent text and metadata, and private classifiers learn from a customer's own taxonomy and tagging.

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.

Under the annual subscription terms, the order's product terms set the license and commercial conditions. Origin can be demonstrated or trialed, and the MAIA assistant is an additional license within it. No standard price appears for either the subscription or the add on, so the offer, renewal and any further product arrangements depend on the contracted rights. Minesoft claims time savings of up to 90% on IP tasks, without a method behind the figure.

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.

Origin serves IP counsel, patent agents, patent offices and company research teams working on prior art, portfolio review, freedom to operate and novelty projects. The data runs to more than 190 million publications from over 110 national and regional authorities, with full text for major jurisdictions and more than 30 years of patent data reviewed daily for quality. Minesoft supports users in more than 15 languages across all time zones and counts more than 90,000 active users across its products.

MAIA adds invalidity, landscape and technology scouting work. The product stays with patent work rather than litigation or case management.

Source: Vendor Published
Sources on file

6 public documents

The public pages on file for Minesoft Origin, 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

No published figure

  • Minesoft Origin is sold on an annual quote.
  • The product has a demo and trial route.
  • MAIA is an additional entitlement within Origin.
  • No standard price appears for either.

The annual contract and product terms determine which services, license rights, subscription term, and fees apply. Origin and PatBase are distinct products. The MAIA assistant requires an additional entitlement.

Implementation: Minesoft provides product support and training resources. No standard implementation or onboarding price for Origin is shown.

Note: The annual terms cover renewals and the commercial terms for each product. The quote can differ by product and by feature entitlements.

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 policy only

Public material states that customer content trains, refines or personalizes models, with no matching term located in the published agreement. Any de identification, anonymization or aggregation qualifier is recorded in the summary.

Under the AI security policy, shared models and MAIA do not train on customer prompts, and user requests and results stay inside Minesoft. A private classification model learns from a customer's behavior and tags without being shared with other users. The annual terms do not address that training distinction.

Source: Vendor Publishedthe model learns from your behaviour and taggingAs 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?

Customer controlled, no zero option

The customer controls the retention window, by product configuration or by contractual instruction, but zero retention is not stated as available.

Users can delete their saved query history at any time. Outside providers behind Doc Chat and summaries keep nothing under zero retention terms. MAIA prompts and generated reports have no stated lifetime inside Minesoft's environment.

Source: Vendor Publishedcan be deleted at any time by the user themselvesAs 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.

Query history and private classifiers stay with the customer that owns them, and MAIA adds role based access and single sign on. A matter permission model for law firms, or retrieval governed by an outside document system's ethical walls, is not part of the product.

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?

Notice committed

Terms commit to notice where lawfully permitted. No transparency report located.

Clause 19 of the annual terms promises as much advance notice as possible of a compelled disclosure of confidential information, where the law permits.

Source: Vendor Publishedit gives the other party as much notice of the disclosure as possibleAs 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.

Origin draws on more than 190 million publications from over 110 national and regional authorities through a live feed reviewed daily for quality, and the retrieval model trains on public patent material. No rights basis is given for each underlying source.

Source: Vendor PublishedAs of Nov 20, 2026Evidence

Good Law Verification

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

Own treatment signal

The vendor computes and surfaces subsequent history itself, with the method described.

MAIA can report legal status and timelines from patent records, and the patent data carries normalized status and family information for triage. A material conclusion on status stays with a patent professional.

Source: Vendor PublishedAs of Nov 20, 2026Evidence

Refusal and Uncertainty Behavior

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

Not addressed

No located public material addresses what the product does when it cannot ground an answer.

Doc Chat works only within the current patent, and Minesoft advises against putting sensitive information into it. MAIA runs with workflow guardrails and human checkpoints. No defined refusal applies when patent records do not support a conclusion.

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 Minesoft Origin. The product finds and compares patent publications and produces patent analyses 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.

Minesoft says MAIA is not a legal expert and that anything legally material needs review by a patent attorney, and MAIA keeps a human checkpoint at each decision. Neither the workflow nor its output is tied to guidance on AI from a named bar or patent regulator.

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.

The platform aims to cut the time practitioners spend finding and comparing prior art, with savings of up to 90% claimed. It offers no billing guidance on AI assisted patent work and produces no attorney fee record for those tasks.

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?

On request only

The material exists behind a sales conversation or an executed agreement.

Minesoft answers 50 to 80 security assessments a year in writing, each signed off by a named person, and describes outside providers for Doc Chat and summaries without naming them. No provider roster for Origin or client disclosure packet is 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.

MAIA exports findings tied to source patents in three report formats and keeps them in an Origin workflow, and its full audit trail records the human checkpoints. No court export combines the model version, the full sources and a record of human verification.

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