Compliance

Open Data

Data made accessible and reusable for public use and research.

How the laws define this

Regulators and policy frameworks agree that open data consists of publicly available, machine-readable datasets accompanied by metadata and licences that allow reuse for innovation, transparency, and research. Most instruments restrict this concept strictly to public-sector or government-produced information, including the Act on Open Data and Re-use of Public Sector Information, the National Data Governance Interim Regulations, the Digital Transformation Bible 2020-2025, and Executive Order 13859. These public-sector frameworks also frequently highlight legal exclusions or safeguards for sensitive data, trade secrets, and privacy. In contrast, FARI defines open data more broadly as any dataset made accessible and managed responsibly for the common good, without limiting its definition to public-sector sources. Similarly, the strategy Toward an AI Strategy in Mexico emphasizes machine processing, broad consumption, and public-sector innovation without embedding specific privacy or classification exclusions directly within its definition.

Synthesised from the 6 statutory definitions below. Each one is quoted in full, with its source.

How jurisdictions differ

What is distinctive in each definition, binding law in force first.

Jurisdiction	Instrument	Status	What is distinctive
United States	Executive Order on Maintaining American Leadership in Artificial Intelligence (American AI Initiative, EO 13859)	In Force	References US federal alignment with OMB Circular A-130 and M-13-13 to guide inventorying, publishing, and improving access to datasets and models with privacy and security safeguards.
Saudi Arabia	National Data Governance Interim Regulations (including Open Data, Data Sharing, Freedom of Information and Personal Data Interim Regulations)	In Force	Limits to government-produced non-sensitive data that must be proactively published in machine-readable formats with metadata and open licenses, excluding sensitive or classified information.
Poland	Act of 11 August 2021 on Open Data and Re-use of Public Sector Information	In Force	Specifies public-sector information must be published without discriminatory conditions or technical barriers via a national portal, subject to restrictions like personal data and trade secrets.
Belgium	FARI AI for the Common Good Institute	In Force	Frames as a non-binding practice and principle for responsibly managed datasets to foster transparency, research, civic innovation, and AI for the common good through collaboration.
Mexico	Toward an AI Strategy in Mexico: Harnessing the AI Revolution	Adopted	Emphasizes standards for interoperability, machine processing, and transparent procurement to unlock public-sector AI innovation and impact assessment.
Greece	Digital Transformation Bible 2020-2025	Adopted	Defines expectations for publication policies enabling analytics and third-party services to promote transparency and innovation.

What it triggers

In Poland, public sector entities must provide high-value and dynamic datasets in open, machine-readable formats via a national portal.source
In Saudi Arabia, government-produced non-sensitive data should be proactively published in machine-readable formats with metadata and an open license.
In the United States, federal agencies must inventory, publish, and improve access to datasets and models while applying privacy and security safeguards.source

Definitions (6)

A practice and principle advocating that datasets (where appropriate) be made accessible, reusable, and managed responsibly to foster transparency, enable research and civic innovation, and support the development of AI for the common good. FARI promotes open data to increase public benefit and facilitate collaboration between researchers, administrations, and citizens.

FARI – AI for the Common Good Institute InitiativeDefinition 1 of 6

Data made publicly available in formats and standards that enable reuse, interoperability, and machine processing, accompanied by metadata and licences that permit broad consumption. The strategy emphasises publishing machine-readable, interoperable datasets to unlock public-sector AI innovation and support transparent procurement and impact assessment.

Hacia una Estrategia de IA en México: Aprovechando la Revolución de la IA (Toward an AI Strategy in Mexico)Definition 2 of 6

Open Data is defined as government-produced non-sensitive data that should be proactively published in machine-readable formats with metadata and an open license permitting reuse, subject to exclusions for legitimately sensitive or classified information.

National Data Governance Interim Regulations (including Open Data, Data Sharing, Freedom of Information and Personal Data Interim Regulations)Definition 3 of 6

Public-sector information published in non-proprietary, machine-readable formats and accompanied by metadata, intended to be freely discoverable and reusable without discriminatory conditions or technical barriers, subject to permitted legal restrictions (e.g., personal data, trade secrets).

Act on Open Data and Re-use of Public Sector Information (Ustawa o otwartych danych i ponownym wykorzystywaniu informacji sektora publicznego)Definition 4 of 6

Data that is made publicly available and structured to be discoverable and reusable, as referenced by the EO in alignment with OMB Circular A-130 and Memorandum M-13-13; the EO uses this concept to guide inventorying, publishing, and improving access to federal datasets and models while applying appropriate privacy and security safeguards.

Executive Order on Maintaining American Leadership in Artificial Intelligence (American AI Initiative, EO 13859)Definition 5 of 6

Public-sector data published in machine-readable formats and under licences or terms that enable reuse, analytics and third‑party services; the strategy defines expectations for publication, metadata, discoverability and reuse policies to promote transparency and innovation.

Digital Transformation Bible 2020-2025 (National digital strategy / National Data Strategy elements)Definition 6 of 6

Not to be confused with

FAIR data

FAIR data focuses on standards making data findable, accessible, interoperable, and reusable, while open data emphasizes public availability and reuse of non-sensitive datasets.

FAIR data principles

FAIR data principles are specific standards for findability, accessibility, interoperability, and reusability, whereas open data centers on public sector publication in machine-readable formats.

Questions and answers

What obligations does open data impose on US federal agencies?

Federal agencies must inventory, publish, and improve access to datasets and models that are publicly available and structured to be discoverable and reusable. Appropriate privacy and security safeguards must be applied. This aligns with OMB Circular A-130 and Memorandum M-13-13.

Does open data require a national portal in Poland?

Yes, Polish public sector entities must provide high-value and dynamic datasets via a national portal in non-proprietary, machine-readable formats with metadata.

What exclusions apply to open data in Saudi Arabia?

Open data is limited to government-produced non-sensitive data, proactively published with metadata and open licenses, excluding legitimately sensitive or classified information.