Q
Qura
Qura is an AI legal research platform from Qura AB of Stockholm, founded in 2024, covering Swedish and EU law. A user describes a question in natural language, and Qura searches more than 1,000 sources, among them case law, legislation and legal literature, through an ontology the company says maps 276 million connections between statutes, recitals, case law and guidance. Qura says it follows a strict zero content generation policy: it returns ranked legal sources with references to the original text rather than generating answers.
Its literature comes partly from data partnerships with the publishers Iustus and Studentlitteratur. Customers named on its site include the law firms Foyen, edpLaw and Bokwall Rislund, and students at several Swedish universities can sign up with a university email. Qura says customer data is never used for AI training, that all data is hosted in the EU, and that it follows the Swedish Bar Association's security and confidentiality demands on data processors.
Legora acquired Qura in April 2026 and said the Qura team would join its legal research organization. Qura's own site continues under its name with a free trial. Pricing is not published.
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.
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 user describes a legal question in natural language, and the platform searches its curated sources and returns ranked results with references to the original text. A post of 29 September 2025 on the legal data ontology counts 276 million connections at chapter, paragraph and word level, linking statutes, recitals, case law and guidance. Because Qura does not generate answers, the models do the finding and ranking rather than the writing.
No workspace, drafting tool or other product is described beside the search. Legora, which acquired Qura in April 2026, announced the deal as building a legal research platform native to AI.
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.
Qura's home page says it uses only curated legal sources, always with clear references to the original text. The output is a ranked list of sources the user opens and reads. The home page also claims that in 27% of searches Qura finds critical sources traditional methods miss, that research is 84% faster, and that 97% of users find research easier. No test set, sample, period or definition is given for any of them. Nothing describes how ranking is evaluated or how often a relevant source is missed.
The FAQ's reliability claim covers the problems of generated text, and says nothing about a search that misses or misranks authority.
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.
Qura's FAQ says the AI does not generate answers: Qura follows a strict zero content generation policy and returns authentic sources. The lawyer reads the original text of each source. A customer interview of 23 October 2025 quotes Bokwall Rislund saying the lawyer always performs the analysis, not the AI model. How ranking decisions are made, what the user is told when coverage of a question is thin, and where language models act in the search are not published.
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.
A post of 9 September 2024 says Foyen rolled Qura out across the firm, nearly 100 practitioners, after a spring beta, and that more than 50 legal teams in the Nordics then used it. A post of 23 October 2025 describes Bokwall Rislund rolling it out across the firm. Qura's home page names Foyen, EQT and edpLaw among its customers, with quotes from a Foyen partner, EQT's interim counsel and an edpLaw partner. The news page carries customer interviews dated from March 2025 to February 2026.
The home page performance figures are tied to no customer, period or method, and no measured result from any named firm is published.
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.
The home page commits to no AI training on customer data, hosting in the EU, and the Swedish Bar Association's demands on data processors. The privacy policy states that it does not cover the input users submit, the outputs or uploaded documents, and no customer terms, data processing agreement or other document governing that material is published. How a firm's connected internal knowledge is separated between users, matters or customers is not described. Privilege and work product are not addressed by name.
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.
Because Qura returns sources rather than answers, it does not produce the conclusion a client would rely on. The about page says Qura combines AI efficiency with human expertise to help lawyers and recommends it to anyone working in law, and students can sign up, so users include people who are not lawyers. No terms of service are published, so no disclaimer, eligibility rule or allocation of responsibility for relying on a search is published. Nothing addresses a lawyer's competence duties when relying on research ranked by AI.
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.
Nothing published sets out how the AI is governed. The about page, the privacy policy, the ontology post and the customer interviews name no person accountable for model behavior, describe no testing before a release, publish no evaluation results and set out no principles for how the AI is built and used. Declining to generate answers is a design choice, and the Swedish Bar Association reference concerns data processing rather than the conduct of the models.
Nothing addresses how ranking performs across areas of law or source types, as the search ranks sources across courts, authorities and publishers and expands beyond Sweden.
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.
The privacy policy, last updated 10 January 2026, sets retention periods for personal data: account data for the life of the account plus one year, billing records six years, and analytics and contracts three years or more. It describes encryption in transit and at rest, and refers to a separate information security policy that is not published. It names no processors. It also states that it does not extend to Own Data, meaning the input users submit, the outputs and uploaded documents.
Nothing published says who processes queries and results or how incidents are handled, and how long that material is kept is left open.
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.
Who bears the loss when the search misses or misranks authority is not addressed. No terms of service, customer agreement or warranty is published, and the footer links only a privacy policy and a cookie policy, both shared as online documents. Qura says searching authentic sources avoids the reliability problems of generated content. No published term covers warranty, indemnity, limitation of liability or recourse. Qura's site does not say that Legora's terms govern it.
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.
Qura's home page says it connects a firm's internal knowledge alongside its sources, without naming a document management system or describing how internal material is brought in. A post of 9 September 2024 describes a partnership with Foytech, under which shared customers use the two products together, with Qura checking Foytech's generative output against legal sources. That pairs two AI tools rather than connecting to a practice system.
No integration with document, matter or knowledge management, no API documentation and no Microsoft Word or Teams add in is published. Nothing on the site describes integration with Legora's platform.
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.
Qura's home page says all data is hosted in the EU, and the privacy policy says all personal data is stored and processed exclusively within the EU and EEA. Qura is delivered as a web platform with its own login. The cloud provider, the country or region within the EU, and whether customers share infrastructure or are separated are not stated. Nor is where language models process queries. The privacy policy's EU statement covers personal data and leaves the queries and documents it calls Own Data outside its scope.
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.
No independent security attestation is published. Qura's home page says it has undergone rigorous security verifications from law firms, banks and government agencies, which describes customer reviews rather than an audit. No ISO 27001 certificate, SOC 2 report, penetration test summary or trust center is named, and the information security policy the privacy policy mentions is not linked from the site.
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 ontology post says Qura uses language models to find analogies across doctrines, recurring legal mechanisms and trends, and the home page refers to AI trained for legal methodology. No model, provider, hosting location for the models or change notice is named, and the privacy policy names no processors. Whether Qura's models now run on Legora's infrastructure after the acquisition is not stated.
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.
No pricing appears at any level. The home page offers a free trial and a demo, and student sign up needs only a university email, with no price, plan, unit of charge or account of what a firm subscription includes. The privacy policy lists subscription details and payment history among the data collected, which points to paid subscriptions without describing them.
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 about page names Sweden as the first market, with EU legal sources, and describes expansion to further jurisdictions as coming. Named users include law firms, corporate legal teams and students at several Swedish universities. The FAQ describes coverage as all legal sources with value, from local case law to international stock market regulations, without saying which areas are thin or which EU materials are included. Jurisdictions beyond Sweden and the EU are not yet covered.
6 public documents
The public pages on file for Qura, 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.
-
quradata.com5 signals
Client Data in Training, Ethical Walls and Matter Segregation, Bar Guidance Alignment and 2 more
Read Oct 2, 2026
-
Prompt and Output Retention, Third Party Request and Subpoena Notice, Outside Counsel Guideline Readiness
Read Oct 2, 2026
-
Fabricated Citation Record
Read Oct 2, 2026
-
quradata.com/about-us1 signal
Refusal and Uncertainty Behavior
Read Oct 2, 2026
-
Good Law Verification
Read Oct 2, 2026
-
Primary Law Corpus Provenance
Read Oct 2, 2026
Legal Signals
What each signal meansA 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.
Client Data in Training
Can material a lawyer puts into this product be used to train a model?
A public policy or trust page states no training on customer content, with no matching term located in the published agreement.
Qura's home page states that customer data will never be used for any type of AI training or fine tuning. No customer agreement or terms of service is published. The privacy policy does not mention training and does not extend to the input, outputs and uploaded documents users submit.
Prompt and Output Retention
How long does the product keep what a lawyer typed, and can that be set to zero?
No located public material states how long prompts and outputs are retained.
Nothing published states how long queries, results or uploaded documents are kept. The privacy policy sets retention periods for personal data such as account and billing records, and expressly excludes the input, outputs and uploaded documents it calls Own Data.
Ethical Walls and Matter Segregation
Does retrieval respect the firm’s ethical walls, or can the model read across them?
No located public material addresses walls or matter level segregation.
Qura's home page says it connects a firm's internal knowledge, and nothing on the home, about or students pages or in the privacy policy describes how that material is kept to the firm, a team or a matter, or separated between customers.
Third Party Request and Subpoena Notice
If someone subpoenas the vendor for a firm’s data, does the firm hear about it first?
No located term or policy addresses third party requests for customer data.
Under the privacy policy, personal data may go to competent authorities when the law requires, with no notice commitment. The policy does not extend to the input, outputs and uploaded documents users submit, and nothing published addresses legal demands for those.
Primary Law Corpus Provenance
Where does the law in this product come from, and does the vendor have the right to use it?
Sources are identified without stating the license or rights basis.
The home page counts more than 1,000 sources, case law, legislation and literature among them, and the ontology post of 29 September 2025 describes 307 data pipelines, APIs and direct court correspondence. Posts of 24 October and 10 November 2025 announce data partnerships with the publishers Iustus, covering literature in more than 25 legal areas, and Studentlitteratur. The terms of those partnerships and the licensing of the other sources are not stated.
Good Law Verification
Does the product tell you when the authority it just cited has been overruled?
No located public material addresses whether authority is checked for subsequent history.
Nothing on the home or about pages or in the ontology post describes a check of whether a returned judgment has been overruled or a provision amended, or a treatment signal on results. The ontology links sources to one another without describing their later history.
Refusal and Uncertainty Behavior
What does the product do when the answer is not in the corpus?
No located public material addresses what the product does when it cannot ground an answer.
Nothing published describes what Qura shows when its sources do not answer a question or coverage is thin. Qura's FAQ says it does not generate answers, which rules out an invented answer without describing an abstention path.
Fabricated Citation Record
Does a public court record exist addressing fabricated or hallucinated legal citations in output from this product?
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.
No court order, opinion or disciplinary record naming Qura as the source of fabricated authority is on the public record. The one entry for Qura in the AI Hallucination Cases database maintained by Damien Charlotin, a Connecticut matter of 31 March 2026, matches a party's surname and does not involve the product.
Bar Guidance Alignment
Has the vendor engaged in public with the ethics opinions its buyers are bound by?
Public materials refer to professional responsibility in general terms without naming guidance.
Qura's home page says it follows the Swedish Bar Association's security and confidentiality demands on data processors, naming the issuing body without naming the guidance or rule. No ethics opinion or professional code provision on AI is named.
Billing and Fee Posture
Does the vendor address what happens to the bill when the work takes an hour instead of six?
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.
Faster research is claimed, at 84% on the home page with no basis stated, and the named users include law firms researching for clients. Nothing published addresses how AI assisted research time is billed or disclosed to a client.
Outside Counsel Guideline Readiness
Can a firm get this vendor through a client’s AI clause without a bespoke negotiation?
No located public material supports a client side disclosure obligation.
No subprocessor list, AI provider disclosure or security documentation for a client's AI terms is published on the home, about or students pages, and the privacy policy names no processors.
Court Disclosure Support
If a judge’s standing order requires an AI disclosure, can the product produce one?
No located public material addresses court disclosure or verification certification.
Nothing on the home or about pages, in the ontology post or in the customer interviews addresses court disclosure of AI assisted research or a record of the searches and sources behind a piece of work.