Regulatory frameworks

AI governance frameworks, standards and regulations covered by Lexora

Organizations face different AI governance expectations depending on their industry, geography, use cases and operating context. This directory explains the 14 regulations, standards and risk-management frameworks that Lexora's product positioning covers, from the EU AI Act to ISO 42001.

Lexora is designed to help organizations manage the governance activities that relate to them. It does not decide which frameworks apply to you.

Framework-based AI governance

Why organizations govern AI against frameworks

Laws, standards and risk-management frameworks give organizations a shared structure for governing AI: what to look at, what to document and who is accountable. The directory below mixes binding regulations, voluntary frameworks and certifiable standards, so each entry states which kind it is.

These themes recur across the frameworks, though each emphasizes them differently.

  • Accountability and oversightClear ownership, and people who can review and intervene.
  • Risk identification and assessmentKnowing what could go wrong, and how serious it is.
  • Policies and proceduresWritten rules that teams can follow and review.
  • Documentation and evidenceRecords that show what was decided and done.
  • Privacy and data protectionRespecting the rules around personal and health data.
  • Management systems and reviewMonitoring, audit and continual improvement over time.

4 frameworks

AI regulation and national AI governance frameworks

Laws, proposed laws and government frameworks written specifically for AI. Their legal force differs, so each entry states its status.

EU AI Act

Binding AI regulation

European Union · Regulation (EU) 2024/1689

The EU AI Act is the European Union's binding law on artificial intelligence. It takes a risk-based approach: certain AI practices are prohibited, high-risk AI systems must meet specific requirements, and some systems carry transparency obligations. Its obligations apply in phases, and it can reach organizations outside the EU whose AI systems are placed on the EU market or whose outputs are used there. Which obligations apply depends on the system and on the organization's role.

Key governance considerations

  • Identifying which AI systems are in scope and their risk category
  • Understanding your role, for example provider or deployer
  • Risk management, documentation and record-keeping for high-risk systems
  • Human oversight and transparency measures
  • Monitoring and incident handling after deployment

How Lexora relates

The deck lists the EU AI Act among the 14 frameworks to which Lexora maps controls. The AI System Registry, Risk Register, Human Oversight workflows, Transparency Portal and Incident Manager can support the registration, oversight and incident-handling work the Act involves. Deciding which category or obligations apply remains the organization's responsibility.

  • AI System Registry
  • Risk Register
  • Human Oversight
  • Transparency Portal
  • Incident Manager

EUR-Lex: Regulation (EU) 2024/1689 (official source, opens in a new tab)

UK AI Framework

AI regulatory principlesNon-statutory

United Kingdom · pro-innovation approach

The UK's pro-innovation approach to AI regulation, set out in a 2023 white paper and the government's February 2024 response, rests on five cross-sectoral principles. They were proposed on a non-statutory basis and are applied by existing regulators, such as the Information Commissioner's Office and the Competition and Markets Authority, within their own remits. For organizations, that means expectations can vary by sector and regulator.

Key governance considerations

  • Safety, security and robustness
  • Appropriate transparency and explainability
  • Fairness
  • Accountability and governance
  • Contestability and redress

How Lexora relates

The deck lists the UK AI Framework among the frameworks Lexora covers. Lexora's Risk Register, Transparency Portal, Human Oversight workflows and Incident Manager can support documentation of how an organization addresses these principles. The principles are not a checklist, and sector regulators decide how they apply.

  • Risk Register
  • Transparency Portal
  • Human Oversight
  • Incident Manager

GOV.UK: government response, February 2024 (official source, opens in a new tab)

Singapore AIGF

AI governance frameworkVoluntary

Singapore · Model AI Governance Framework

Singapore's Model AI Governance Framework, issued by the Infocomm Media Development Authority (IMDA) and the Personal Data Protection Commission (PDPC), is voluntary guidance for organizations that deploy AI. Its second edition, released in January 2020, offers practical measures in four areas and includes industry examples. It is often used as a practical reference for internal AI governance even outside Singapore.

Key governance considerations

  • Internal governance structures and measures
  • Determining the level of human involvement in AI-augmented decision-making
  • Operations management
  • Stakeholder interaction and communication

How Lexora relates

The deck lists Singapore AIGF among Lexora's 14 frameworks. Policy generation, Human Oversight workflows, the Risk Register and the Transparency Portal can support the internal governance, human-involvement and communication measures the framework describes. Adoption of the framework is voluntary, and Lexora does not assess conformance to it.

  • Policy generation
  • Human Oversight
  • Risk Register
  • Transparency Portal

PDPC: Model AI Governance Framework, second edition (PDF) (official source, opens in a new tab)

Canada AIDA

AI & data actProposed · not in force

Canada · Artificial Intelligence and Data Act

The Artificial Intelligence and Data Act (AIDA) was proposed Canadian federal legislation, introduced as part of the Digital Charter Implementation Act, 2022 (Bill C-27). It would have set requirements for the design, development and deployment of AI systems, scaled to risk, with particular attention to high-impact systems. Bill C-27 did not receive Royal Assent before the first session of the 44th Parliament ended on 6 January 2025, and the Government of Canada's AIDA page is archived. AIDA is not in force; organizations should check for any successor legislation.

Key governance considerations

  • Identifying risks of harm and bias at the design stage
  • Assessing intended uses and limitations during development
  • Mitigating risk and monitoring continuously at deployment
  • Watching for successor legislation

How Lexora relates

The deck lists Canada AIDA among the 14 frameworks. Because AIDA was proposed and is not in force, it is best read as a reference point for risk-based governance. Lexora's Impact Assessment, Risk Register and Incident Manager can support the identification, mitigation and monitoring work its proposals described.

  • Impact Assessment
  • Risk Register
  • Incident Manager

Government of Canada: AIDA (archived page) (official source, opens in a new tab)

5 frameworks

AI standards and risk-management frameworks

Voluntary standards and frameworks that describe how to manage AI systems and their risks. Some, such as ISO/IEC 42001, can be assessed by independent certification bodies.

ISO 42001

AI management system standard

International · ISO/IEC 42001:2023

ISO/IEC 42001:2023 is the international standard for an AI management system. It sets out requirements for establishing, implementing, maintaining and continually improving a structured approach to managing AI responsibly within an organization that provides or uses AI-based products or services. Because it is a management-system standard, an organization's AI management system can be assessed by an independent certification body.

Key governance considerations

  • Defining an AI policy, roles and responsibilities
  • Assessing AI risks and the impact on individuals and society
  • Applying controls across the AI system lifecycle
  • Managing suppliers and other third parties
  • Internal audit, management review and continual improvement

How Lexora relates

The deck lists ISO 42001 among the frameworks Lexora covers. The Controls Library, Policy generation (from 42 pre-built templates), Gap Analysis and Maturity Assessment can support the policy, control and improvement work an AI management system involves. Lexora does not certify organizations; certification is carried out by independent certification bodies.

  • Controls Library
  • Policy generation
  • Gap Analysis
  • Maturity Assessment

ISO: ISO/IEC 42001:2023 (official source, opens in a new tab)

NIST AI RMF

AI risk management frameworkVoluntary

United States · AI RMF 1.0 (NIST)

The NIST AI Risk Management Framework (AI RMF 1.0) was published by the US National Institute of Standards and Technology in January 2023. It is voluntary and sector-agnostic, designed to help organizations build trustworthiness considerations into the design, development, use and evaluation of AI systems. It is organized around four functions: Govern, Map, Measure and Manage.

Key governance considerations

  • Govern: policies, accountability and risk culture
  • Map: understanding context, intended use and potential impacts
  • Measure: analysing and tracking identified risks
  • Manage: prioritizing and responding to risks over time

How Lexora relates

NIST AI RMF is on the deck's list of 14 frameworks. The Risk Register, Impact Assessment and Risk Traceability can help teams record and follow the risks that the Map, Measure and Manage functions address. The framework is voluntary, and using Lexora does not by itself make an organization aligned with it.

  • Risk Register
  • Impact Assessment
  • Risk Traceability
  • Maturity Assessment

NIST: AI Risk Management Framework (official source, opens in a new tab)

ISO 23894

AI risk management guidance

International · ISO/IEC 23894:2023

ISO/IEC 23894:2023 provides guidance on managing risk specific to AI. It adapts the general risk-management approach of ISO 31000 and describes how to integrate AI risk management into an organization's activities across the AI system lifecycle. It is guidance rather than a requirements standard, so organizations use it to shape their own risk processes.

Key governance considerations

  • Embedding AI risk management in organizational processes
  • Establishing scope, context and risk criteria
  • Identifying, analysing and evaluating AI risks
  • Treating risks and recording the decisions taken
  • Monitoring, review and communication

How Lexora relates

The deck lists ISO 23894 among Lexora's 14 frameworks. The Risk Register, Risk Traceability (following a risk from identification to resolution), Impact Assessment and Remediation Tracker can support the identification, treatment and monitoring steps the guidance describes.

  • Risk Register
  • Risk Traceability
  • Impact Assessment
  • Remediation Tracker

ISO: ISO/IEC 23894:2023 (official source, opens in a new tab)

ISO 22989

AI concepts & terminology

International · ISO/IEC 22989:2022

ISO/IEC 22989:2022 establishes terminology for artificial intelligence and describes concepts in the field, including AI system lifecycle stages and stakeholder roles. It is a foundation standard rather than a management-system standard. Its value in governance is a shared vocabulary, so legal, risk, engineering and business teams can describe AI systems in the same terms.

Key governance considerations

  • Agreeing common definitions across teams
  • Describing AI systems and lifecycle stages consistently
  • Identifying stakeholder roles across the AI value chain
  • Using consistent terms in policies, risk records and reports

How Lexora relates

ISO 22989 is among the 14 frameworks on the deck. A shared vocabulary helps when teams record AI systems consistently in the AI System Registry and write policies and risk records that others can read. The deck does not describe a mapping specific to this standard beyond Lexora's general mapping of controls to frameworks.

  • AI System Registry
  • Knowledge Base

ISO: ISO/IEC 22989:2022 (official source, opens in a new tab)

IEEE 7000

Ethical AI design standard

International · IEEE 7000-2021

IEEE 7000-2021 is the IEEE Standard Model Process for Addressing Ethical Concerns during System Design. It describes a process for identifying stakeholders and their values, turning those values into system requirements, and tracing them through risk-based design decisions. It applies to the engineering of systems in general, which makes it relevant to AI systems where concerns such as fairness, privacy and transparency should be considered early.

Key governance considerations

  • Identifying stakeholders and eliciting their values
  • Translating ethical values into system requirements
  • Making risk-based design decisions and documenting the reasoning
  • Tracing values through to design and review

How Lexora relates

IEEE 7000 is on the deck's list. Impact Assessment (evaluating AI system impact before deployment), Risk Traceability and Human Oversight workflows can support documenting ethical considerations and the decisions taken about them. Lexora does not provide the engineering process the standard describes.

  • Impact Assessment
  • Risk Traceability
  • Human Oversight

IEEE Standards Association: IEEE 7000-2021 (official source, opens in a new tab)

3 frameworks

Privacy and data protection

Laws that govern how personal and health information may be collected, used and protected, including when AI systems process it.

GDPR

Data protection regulation

European Union · Regulation (EU) 2016/679

The General Data Protection Regulation governs the processing of personal data of individuals in the EU and has applied since May 2018. It matters for AI because many AI systems are trained on, or make decisions about, personal data. It sets principles such as lawfulness, fairness and transparency, gives individuals rights over their data, and includes provisions on automated decision-making and data protection impact assessments.

Key governance considerations

  • Establishing a lawful basis and meeting transparency duties
  • Data protection impact assessments where processing is likely to result in high risk
  • Safeguards around solely automated decisions with significant effects
  • Supporting individuals' rights requests
  • Accountability and records of processing

How Lexora relates

GDPR is on the deck's list. Lexora includes automated privacy impact assessments (DPIA) and Gap Analysis, which can support privacy governance work for AI systems. Lexora does not decide whether a lawful basis exists or give legal advice; those judgements stay with the organization and its advisers.

  • DPIA
  • Gap Analysis
  • Controls Library
  • Knowledge Base

EUR-Lex: Regulation (EU) 2016/679 (official source, opens in a new tab)

CCPA

Consumer privacy law

United States · California

The California Consumer Privacy Act (CCPA), as amended by the California Privacy Rights Act, gives California residents rights over the personal information that businesses collect, including the rights to know, delete, correct and opt out of the sale or sharing of their data. AI systems that use personal information fall within its scope where the law applies. California's privacy regulator has also been developing rules on automated decision-making, so organizations should check the current requirements.

Key governance considerations

  • Working out whether and how the law applies to your business
  • Notice at collection and privacy policy content
  • Handling consumer rights requests
  • Limits on sale, sharing and use of sensitive personal information
  • Service-provider and contractor terms

How Lexora relates

The deck lists CCPA among the 14 frameworks, described as consumer privacy law. Gap Analysis, the Controls Library, the Knowledge Base and Reg Deadlines can help teams organize privacy governance work and track relevant dates. Whether the CCPA applies to a given business is a legal question for the organization to settle.

  • Gap Analysis
  • Controls Library
  • Knowledge Base
  • Reg Deadlines

California Attorney General: CCPA (official source, opens in a new tab)

HIPAA

Healthcare data protection

United States · federal law

The Health Insurance Portability and Accountability Act (HIPAA) is a US federal law. Its Privacy, Security and Breach Notification Rules protect individually identifiable health information held by covered entities, such as health plans and healthcare providers, and by their business associates. It is relevant to AI when systems process protected health information, for example in clinical, administrative or analytics settings.

Key governance considerations

  • Risk analysis and risk management for systems handling protected health information
  • Administrative, physical and technical safeguards
  • Business associate agreements with vendors
  • Breach notification and incident response

How Lexora relates

The deck lists HIPAA among the 14 frameworks, described as healthcare data protection. The Controls Library, Gap Analysis, Risk Register, Incident Manager and third-party risk (TPRM) capabilities can support safeguards, vendor oversight and incident tracking. Lexora does not determine whether an organization is a covered entity or business associate.

  • Controls Library
  • Gap Analysis
  • Risk Register
  • Incident Manager
  • Third-party risk (TPRM)

US HHS: HIPAA for professionals and individuals (official source, opens in a new tab)

2 frameworks

Information security and assurance

Standards and reporting frameworks that organizations use to show how information is protected, which AI systems depend on.

ISO 27001

Information security standard

International · ISO/IEC 27001:2022

ISO/IEC 27001:2022 is the international standard for an information security management system. It specifies requirements for establishing, implementing, maintaining and improving a systematic, risk-based approach to protecting the confidentiality, integrity and availability of information, and organizations can be certified by accredited certification bodies. It is relevant to AI governance because AI systems depend on the data, infrastructure and access controls that such a system governs.

Key governance considerations

  • Defining the scope and security objectives
  • Information security risk assessment and treatment
  • Selecting controls and recording a statement of applicability
  • Internal audit and management review
  • Continual improvement

How Lexora relates

The deck lists ISO 27001 among the frameworks Lexora covers. The Controls Library, Gap Analysis, Maturity Assessment, Incident Manager and Remediation Tracker can support an organization's own security governance work. Lexora itself is not yet independently certified (Lexora Security Overview v1.0, October 2026: SOC 2 / ISO 27001 target: Q4). Microsoft Azure holds ISO/IEC 27001 certification and SOC 2 reports for its infrastructure; those are Microsoft's, not Lexora's.

  • Controls Library
  • Gap Analysis
  • Maturity Assessment
  • Incident Manager
  • Remediation Tracker

ISO: ISO/IEC 27001 (official source, opens in a new tab)

SOC 2

Security & trust criteria

United States · AICPA Trust Services Criteria

SOC 2 is an AICPA reporting framework in which an independent CPA firm examines a service organization's controls against the Trust Services Criteria: security, availability, processing integrity, confidentiality and privacy. It is not a law or an AI-specific standard, but customers often request SOC 2 reports from software and AI vendors as assurance about how data is protected. Reports are issued by auditors; an organization cannot self-certify.

Key governance considerations

  • Scoping which Trust Services Criteria apply
  • Documenting controls and keeping evidence that they operate
  • Managing vendors and monitoring control performance over time
  • Preparing for a Type 1 (point in time) or Type 2 (period of time) examination

How Lexora relates

The deck lists SOC 2 among the 14 frameworks, described as security and trust criteria. The Controls Library, Gap Analysis, Remediation Tracker and Risk Traceability (which links controls and evidence) can support the control documentation and follow-up work. Lexora itself is not yet independently certified (Lexora Security Overview v1.0, October 2026: SOC 2 / ISO 27001 target: Q4). Microsoft Azure holds ISO/IEC 27001 certification and SOC 2 reports for its infrastructure; those are Microsoft's, not Lexora's.

  • Controls Library
  • Gap Analysis
  • Remediation Tracker
  • Risk Traceability

AICPA & CIMA: SOC suite of services (official source, opens in a new tab)

Frameworks and the Lexora platform

From what a framework asks to where the work gets done

A framework describes requirements, principles or practices. Teams then do the governance work, and Lexora gives that work a structured home.

  1. What the framework describes

    Requirements, principles or good practice, from a regulator, standards body or government.

  2. What your teams do

    Inventory AI systems, assess risk, select and document controls, keep evidence, handle incidents and report to leadership.

  3. Where Lexora fits

    Lexora maps controls to the frameworks you are accountable to, so a single piece of evidence can support multiple audits at once. You can map each system to the 14 frameworks, run gap analysis and DPIAs, and surface prioritized remediation.

See how the Lexora platform works

Frameworks and compliance modules

A framework and a compliance module are not the same thing

A framework such as ISO 42001 sets out what an organization is expected to consider. A compliance module is part of the platform, such as the Risk Register, Controls Library or Gap Analysis, that helps carry out related work. Lexora organizes its functionality into 34 compliance modules that are integrated rather than separate tools.

Explore the compliance modules

compliance modules, fully integrated

Choosing relevant frameworks

Which frameworks matter depends on your circumstances

Not every framework applies to every organization. These are general considerations, not legal advice and not an assessment Lexora performs for you. Organizations should determine which requirements apply to them and seek qualified advice where needed.

  • Jurisdiction and applicable lawWhere you operate, and whose residents or markets you serve.
  • Industry and use caseFor example, healthcare data or consumer-facing services.
  • Types of AI systemsWhat you build, buy or deploy, and how it is used.
  • Personal dataWhether AI systems process personal or health information.
  • Risk profileThe potential impact of your systems on people.
  • Policies and contractsCustomer, partner and internal commitments, such as requests for SOC 2 reports.

FAQ

Frequently asked questions about AI governance frameworks

What is the difference between an AI regulation and an AI governance framework?

A regulation is a law that organizations must follow where it applies, such as the EU AI Act. Standards and frameworks, such as ISO 42001 or the NIST AI RMF, are voluntary guidance or specifications; some, like ISO 42001, can be assessed by independent certification bodies. Others sit in between: the UK principles are non-statutory, and Singapore's framework is voluntary. This directory includes all of these, and each entry states its type.

Do all 14 frameworks apply to every organization?

No. Relevance depends on jurisdiction, industry, the AI systems you use and the data they process. Lexora does not determine an organization's legal obligations. Decide which frameworks apply with your own legal and compliance advisers, then use Lexora to organize the related governance work.

How can an AI governance platform help with framework-related activities?

Framework work involves recurring activities: keeping an inventory of AI systems, assessing risk, selecting and documenting controls, keeping evidence, handling incidents and reporting to leadership. Lexora maps controls to the 14 frameworks, so a single piece of evidence can support multiple audits at once, and provides modules for these activities. See the platform overview.

Does using Lexora guarantee compliance?

No. Lexora supports governance activities; it does not certify organizations, give legal advice, replace independent audits or remove an organization's own accountability. Lexora itself is not yet independently certified.

How do frameworks relate to compliance modules?

A framework describes requirements, principles or practices. A compliance module is a functional part of Lexora, such as the Risk Register or Gap Analysis, that helps an organization carry out related work. Lexora organizes its functionality into 34 compliance modules. A dedicated page on compliance modules is planned.

Is Canada's AIDA in force?

No. AIDA was proposed as part of Bill C-27, which did not receive Royal Assent before the first session of the 44th Parliament ended on 6 January 2025. Check official sources for any successor legislation.

See how Lexora supports your AI governance work

Talk to our sales team about the frameworks that matter to your organization and how Lexora can help you manage the related work.

Request a Demo