Trustworthy AI

A conceptual standard for AI systems that meet legal requirements, adhere to ethical norms (including fairness and non-discrimination), and demonstrate technical robustness and safety; intended as the guiding principle for national guidelines and alignment with EU frameworks.

An operational requirement that AI system design choices adhere to principles of explainability, fairness, transparency, and energy/resource efficiency consistent with evolving European ethical guidance and regulatory direction; applicants must describe how these considerations are integrated into system design and project deliverables.

The strategic principle guiding AIM AT 2030 that requires AI development and deployment to respect fundamental rights, fairness, transparency, accountability and security; it frames ethical, legal and social safeguards and aligns national actions with EU norms on trustworthy AI.

AI systems designed and deployed in adherence to human-centred values, legal requirements and ethical standards, with the aim of protecting fundamental and human rights, ensuring fairness, inclusion and respect for societal requirements.

AI designed and implemented in accordance with ethical principles including human oversight, fairness, transparency and robustness; presented as the normative objective guiding assessments, procurement guidance and adoption support within AI4Belgium.

AI systems that comply with applicable law, align with ethical principles (e.g., human dignity, autonomy, prevention of harm, fairness, explicability), and demonstrate technical robustness (reliability, safety and resilience) across their lifecycle. The Guidelines frame Trustworthy AI as an integrated objective requiring organisational governance, documentation, risk assessment and ongoing monitoring throughout design, deployment and maintenance.

AI that meets the seven ALTAI requirements — lawful, ethical and robust — encompassing human agency and oversight, technical robustness and safety, privacy and data governance, transparency, diversity/non-discrimination and fairness, societal and environmental well‑being, and accountability. The term denotes systems designed, deployed and monitored to align with legal and ethical expectations throughout their lifecycle.

AI systems that comply with applicable law, respect fundamental rights and ethical principles, and are technically robust and secure. The Plan uses this term to frame deployments that must be lawful, ethical and resilient to risks in order to gain public trust.

Defined as AI designed and deployed with respect for fundamental rights, privacy, transparency, and technical robustness; aligning with EU principles for ethical and lawful AI. The Programme adopts this operational concept to guide funding, evaluation, certification and alignment with forthcoming EU regulatory instruments.

AI systems and practices that adhere to principles protecting human rights, privacy (including GDPR compliance), fairness, explainability, safety and security; the concept underpins ethics-by-design, impact assessments and documentation/auditability requirements recommended by the strategy.

Adopted from OECD and EU principles, 'trustworthy AI' denotes systems designed and deployed in ways that respect fundamental rights, ensure transparency and explainability, maintain safety and robustness, and enable accountability and redress. The policy uses this concept to frame ethics frameworks, privacy safeguards, and conformity approaches for public and sectoral deployments.

AI systems that combine safety and security with respect for human rights, transparency of operation, and mechanisms for accountability and oversight. Trustworthy AI requires clear governance, documentation, and stakeholder engagement so users and affected parties can understand and challenge outcomes.

AI systems that satisfy established benchmarks and criteria across safety, fairness, privacy, transparency, and accountability, such that they are considered reliable and appropriate for deployment under the AIEC evaluation framework.

An operational term used in the Update to denote AI systems designed and deployed in ways that comply with data protection rules, uphold fundamental rights, and follow ethical principles (e.g., human-centred design, transparency and societal acceptance). The Update frames funding and projects to promote such systems.

An operational set of principles emphasising human-centred design, transparency and robustness for AI systems; used in the Action Plan as a guiding normative framework for research, infrastructure and application development while deferring to statutory definitions at EU level. The term frames ethical, social and technical safeguards to be advanced through multidisciplinary research and stakeholder dialogue.

AI systems and practices that are designed, developed, and deployed in ways that respect human rights and democratic values, promoting fairness, non-discrimination, transparency, and accountability. The term is used as the guiding principle for GPAI's workstreams and recommendations.

A working concept adopted in the SAPAI describing AI systems designed, developed and deployed in line with principles such as legality, transparency, human oversight, privacy protection and non‑discrimination; used to guide ethics guidance, fairness and bias mitigation, privacy-preserving techniques, and related legal research.

Trustworthy AI denotes AI development and deployment guided by articulated ethical principles — including transparency, human oversight, privacy protection, explainability and cautious testing — ensuring that systems respect fundamental rights and public trust throughout their lifecycle.

'Trustworthy AI' and 'algorithmic systems' are described by the strategy via principles requiring transparency, human oversight, and compliance with privacy and data-protection obligations, with practical guidance to ensure ethical use of AI in public services.

AI systems that are reliable, transparent, explainable, and subject to accountability and oversight mechanisms so that stakeholders can understand, challenge, and trust system behavior and decision-making.

Characteristics of AI systems that include being valid and reliable, safe, secure and resilient, accountable and transparent, explainable and interpretable, privacy-enhanced, and fair with harmful bias managed. Achieving trustworthiness is crucial for public acceptance and effective deployment of AI technologies.

AI systems that are reliable, secure, robust, and operate in a manner that inspires confidence among users and the broader public, built upon a foundation of ethical principles and robust governance frameworks.

AI systems that are developed and used in accordance with ethical principles, legal requirements, and technical robustness standards.

AI systems that are safe, secure, and developed in alignment with democratic values, human rights, and the rule of law.

Artificial intelligence that is designed and operated to be reliable, safe, fair, and transparent, ensuring alignment with the rule of law and human-centric values.

AI systems designed, developed, and deployed in a manner that is safe, human-centric, responsible, and respects human rights and democratic values, fostering public trust and confidence.

Encompasses AI systems designed to be safe, secure, resilient, transparent, explainable, interpretable, privacy-enhanced, fair, and accountable. This multi-faceted definition serves as a guiding principle for all AI development and deployment under the national policy framework, ensuring ethical and reliable operation.

AI systems that are developed and deployed in a manner consistent with the values-based principles outlined in the Recommendation, ensuring they are beneficial, human-centred, robust, secure, safe, transparent, explainable, and accountable.

Trustworthy AI refers to AI systems that embody characteristics such as validity, reliability, safety, security, resilience, accountability, transparency, explainability, interpretability, privacy-enhancement, and fairness, with harmful biases actively managed. These attributes are fundamental to minimizing unintended negative outcomes.

Trustworthy AI encompasses a set of characteristics crucial for AI systems, especially in high-stakes environments like critical infrastructure. These characteristics include validity and reliability, safety, security, resilience, accountability, transparency, explainability, interpretability, privacy, and fairness. Ensuring these attributes helps build confidence in AI systems and minimizes adverse impacts, making them suitable for sensitive applications.

An overarching principle guiding the development and deployment of artificial intelligence systems that are reliable, safe, secure, transparent, fair, and accountable. The concept emphasizes ethical considerations, bias mitigation, and human-centric design, ensuring that AI technologies serve societal good and maintain public confidence. The Workforce for AI Trust Act specifically aims to cultivate a workforce capable of building, evaluating, and governing such systems.

Trustworthy AI refers to artificial intelligence systems that are developed and deployed in a manner that ensures reliability, safety, and adherence to ethical principles. This concept often encompasses aspects like fairness, transparency, robustness, and privacy, aiming to build public confidence and mitigate potential harms.