Legal sources with official primary sources

Printed ·

SMLM Previewerrors, bugs and incomplete coverage may occurThe model is production-ready during Q4-26

Skip to main content
Skip to the answer

About NovaCopilot

NovaCopilot transforms public legal, regulatory and supervisory information into continuously updated business outcome intelligence. With a focus on obligations, controls, risks, consequences and opportunities, organisations gain source-linked evidence for decisions across markets, industries and jurisdictions.

From public information to business decisions

Regulation is the starting point, not the end product. NovaCopilot structures source material into answers and objects that organisations can use in their own services and workflows.

Businesses and service providers build their own documents, spreadsheets, presentations and services on NovaCopilot decision output in JSON. Attribution remains Source: NovaCopilot.

Sources, agents and knowledge graphs

Source Multi-Language Model (SMLM), regulatory knowledge graphs and intelligence agents structure source-linked information on what applies, what action is required and which consequences the sources support. SMLM is in preview, with explicit limitations.

Rule-based objects can be reproduced from the same register version and inputs. Optional reasoning uses a language model with evidence from the register and is distinct from fixed objects.

Sources and changes

Evidence comes from registered statutes, legal acts, judgments and supervisory decisions. Official sources take priority. Source links, content dates and verification status show the available evidence.

Content follows source-linked changes. A new check or publication date does not, by itself, mean the law or answer has changed.

Limits

NovaCopilot does not replace legal advice or guarantee business outcomes. Missing sources and connections are disclosed, with not_in_register in machine-readable answers.

Question answering, document analysis and account features have different processing paths. Do not include sensitive personal data or confidential material in a question.

The Swedish source register

Statutes
52
Decisions
12713
Languages
4 (SV, EN, NO, DE)
Register engine version
legal-answer/1.1.0
Case law read
2026-09-05
Language model in fixed register lookup
None in fixed register lookup

The people behind the engine

The register is built and reviewed by named people. Here you can see who is responsible for what, and verify the profiles directly.

Andreas Bryntesson

Co-founder of ExploreWorldAI

Technical lead and co-founder. Responsible for the platform, the data engine and daily operations.

LinkedIn

Rebecca Rauma

Head of Sales

Head of Sales. Responsible for customer relationships, partnerships and commercial expansion.

LinkedIn

Daniel Bramme

Co-founder of Valkiv Ventures AB

Co-founder of Valkiv Ventures AB, the company behind the service.

LinkedIn

Named sources with read dates

Every row in the engine comes from one of the registers below. The date is the day the register was read into the snapshot the answers are built against.

Register and publisherWhat is readRowsRead
Svensk författningssamling (SFS)Sveriges riksdagStatute identifier, SFS number, official title and entry into force.52
Riksdagens dokumentregister (propositioner, SOU, Ds)Sveriges riksdagPreparatory work reference and title, linked to the right statute.75
Myndigheternas föreskriftssamlingarIMYFS, FFFS, MSBFS, PTSFS m.fl.Regulation reference and title from each agency's own collection.12
Vägledande avgöranden, HD och HFDrattspraxis.etjanst.domstol.seCase reference, court, headnote summary and the statutes applied.12713
EUR-LexEuropeiska unionens publikationsbyråEU act identifier, CELEX number, entry into force and date of application.32

Personal data handling

What is processed, on what basis and for how long.

  • The controller is Valkiv Ventures AB, Kungsgatan 8, 111 43 Stockholm, Sweden. Org.nr 556995-1311. VAT SE556995131101.
  • A question is matched against source-linked records. The first document analysis reads the document in the browser and matches it against the register without calling a model.
  • Question text may be stored without an account or device link, with email addresses, phone numbers and personal or company identifiers masked. Operational logs support availability and abuse prevention. Processing is described in the privacy policy.
  • Rule-based objects can be reproduced from the same register version and inputs. Optional reasoning uses a language model with evidence from the register and is distinct from fixed objects.
  • The legal basis for operational logs is legitimate interest under Article 6(1)(f) of Regulation (EU) 2016/679: keeping the service available and preventing abuse.
  • Data subjects may request access, rectification and erasure, and may lodge a complaint with the Swedish Authority for Privacy Protection.

Report a deviation

If a row points at the wrong source, a date is off, or a decision has been overruled, write to us. Every report is answered, and a confirmed deviation is corrected in the register and written into the version history.

Reports go to the same address as data subject requests. We confirm receipt within two working days.

  • The URL of the row concerned
  • What is wrong
  • The primary source that shows the correct position

Write to daniel@valkiv.com

Verifiable trust signals

  • Source-linked answers and explicit evidence gaps
  • Obligations, controls, risks and consequences
  • Stable objects and verifiable checksums
  • Rule-based objects, optional source-linked reasoning
  • Model card, source status and audit

Model cardAudit