Trustworthiness
Ethics, safety, transparency and fundamental-rights protections for AI.
How the laws define this
Regulators and standards bodies agree that trustworthiness requires an AI system to consistently deliver safety, privacy, and security while instilling stakeholder confidence and public trust through accountable, transparent performance. These frameworks consistently treat the term as a multi-faceted objective incorporating overlapping safeguards like reliability and risk management. Their primary divergence lies in whether trustworthiness is anchored in human rights and law or in technical operational controls. The Coordinated Plan on Artificial Intelligence (EU, 2021 review) defines it as a policy concept centered on ethical principles, lawful deployment, and respect for fundamental rights. In contrast, technical standards define it through system capabilities and operational mechanisms: ISO/IEC 22989:2022 focuses on the verifiable ability of a system to meet stakeholder expectations, while DKS 3007:2024 frames it as a composite quality established through organizational, technical, and lifecycle controls to enable controllability, explainability, and human oversight.
Synthesised from the 3 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. Showing 2 of the 3 instruments that define this term; all are listed under Definitions below.
| Jurisdiction | Instrument | Status | What is distinctive |
|---|---|---|---|
| Kenya | DKS 3007:2024 Information technology — Artificial Intelligence — Code of Practice for AI Applications | Adopted | Adds controllability, explainability and human oversight as key qualities, framing trustworthiness as the objective achieved through organisational, technical and lifecycle controls. |
| Bulgaria | Coordinated Plan on Artificial Intelligence (EU, 2021 review) — (applies to Bulgaria as EU Member State) | Adopted | Treats trustworthiness as a policy concept emphasising compliance with ethical principles, accountability and fundamental rights, promoting mainstreaming of guidelines, risk assessments and safety measures. |
Definitions (3)
A policy concept framing AI systems’ compliance with ethical principles, safety requirements, transparency, accountability, and respect for fundamental rights. The Plan promotes mainstreaming ethical guidelines, risk assessments, safety measures and transparency practices to build public trust and ensure lawful, rights-respecting AI deployment.
A composite quality of an AI system covering transparency, explainability, controllability, reliability, safety, security and privacy that enables appropriate human oversight and stakeholder confidence. The Code frames trustworthiness as the objective of organisational, technical and lifecycle controls to ensure predictable, accountable and acceptable AI behaviour.
The ability of an AI system to meet stakeholder expectations in a verifiable way, encompassing safety, security, privacy, and reliability.
Not to be confused with
Trusted/Trustworthy AI refers specifically to systems designed to respect human rights, fairness, transparency and safety, while trustworthiness here describes broader abilities, qualities or policy compliance.
Trustworthy/ethical AI emphasises design with transparency, accountability and rights respect, whereas trustworthiness encompasses verifiable performance qualities or policy frameworks beyond just ethical design.
Questions and answers
What does ISO say trustworthiness in AI means?
ISO/IEC 22989:2022 defines it as the ability of an AI system to meet stakeholder expectations in a verifiable way. This covers safety, security, privacy, and reliability. It establishes common terminology for AI governance.
How does Kenya's AI code frame trustworthiness?
DKS 3007:2024 describes it as a composite quality including transparency, explainability, controllability, reliability, safety, security and privacy. It enables human oversight and stakeholder confidence through lifecycle controls.
What is trustworthiness under EU's Coordinated Plan?
The 2021 Coordinated Plan frames it as compliance with ethical principles, safety, transparency, accountability and fundamental rights. It promotes guidelines, risk assessments and transparency to build trust.
Related Terms
Trusted/Trustworthy AI
Systems designed to respect human rights, fairness, transparency and safety....
human‑centred / trustworthy AI
Principles ensuring rights, transparency and safety in AI use....
Trust/security
Privacy, data governance, robustness and cybersecurity....
trustworthy/ethical AI
AI designed with transparency, accountability and respect for rights....
trusted AI
AI designed with safety, explainability, fairness and resilience....