Ethics Guidelines for Trustworthy AI

European Commission

Policy type icon
Policy guidance
Alternative names icon
Private and public sector

Policy overview

The Ethics Guidelines for Trustworthy AI proposed by the High-Level Expert Group on AI, set out a framework for achieving trustworthy AI. The Ethics Guidelines for Trustworthy AI were created to support the implementation of the European Commission's vision for AI. In its vision for AI, the Commission has emphasised the development of ethical, secure, and cutting-edge AI within Europe. The Guidelines outline three components for trustworthy AI that must be upheld throughout the entire lifecycle of the system: firstly, it must be lawful, complying with all relevant regulations; secondly, it must be ethical by adhering to ethical principles and values; and thirdly, it should demonstrate robustness, considering both technical and social aspects to prevent unintentional harm, even when operating with good intentions. The Guidelines were introduced on April 8, 2019, not just as a list of ethical principles but with the intention of providing practical guidance on how such principles can be operationalised in sociotechnical systems.

The Guidelines consists of non-biding best practice and they are addressed to all AI stakeholders. The AI stakeholders encompass actors designing, developing, deploying, implementing, using or being affected by AI, including but not limited to companies, organisations, researchers, public services, government agencies, institutions, civil society organisations, individuals, workers and consumers.

The Guidelines put forward a set of seven key requirements that AI systems should meet in order to be deemed trustworthy. These include the promotion of human agency and oversight, ensuring technical robustness and safety, upholding privacy and data governance, fostering transparency, preventing bias and promoting fairness, considering societal and environmental well-being, and establishing mechanisms for accountability. To assess compliance with these requirements, a specific assessment checklist is provided. These guidelines aim to ensure that AI systems empower individuals, are secure and reliable, respect privacy, operate transparently, avoid bias, benefit society, and have accountability measures in place.

Templates in Saidot

No items found.