The EU Approach to Ethics Guidelines for Trustworthy Artificial Intelligence

By Nathalie A. Smuha. Computer Law Review International, 2019

345 citations recorded by the source corpus.

A continuous journey towards an appropriate governance framework for AI As part of its European strategy for Artificial Intelligence (AI), and as a response to the increasing ethical questions raised by this technology, the European Commission established an independent High-Level Expert Group on Artificial Intelligence (AI HLEG) in June 2018. The group was tasked to draft two deliver-ables: AI Ethics Guidelines and Policy and Investment Recommendations. Nine months later, its first deliverable was published, putting forward a comprehensive framework to achieve “ Trustworthy AI ” by offering ethical guidance to AI practitioners. This paper dives into the work carried out by the group, focusing in particular on its AI Ethics Guidelines. First, this paper clarifies the context that led to the creation of the AI HLEG and its mandate (I.). Subsequently, it elaborates on the Guidelines ’ aim and purpose (II.), and analyses the Guidelines ’ drafting process (III.). Particular focus is given to the questions surrounding the respective role played by ethics and law in the AI governance landscape (IV.), as well as some of the challenges that had to be overcome throughout the process (V.). Finally, this paper places the Guidelines in an international context, and sets out the next steps (VI.) ahead on the journey towards an appropriate governance framework for AI (VII.).

Subfields: Artificial Intelligence, Health Informatics, Safety Research

View Paper: https://www.semanticscholar.org/paper/a00ee7ae4642b986b622e0f48c845e79707b707b

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Health Informatics

In Safety Research

Citations in the corpus

Trustworthy Artificial Intelligence: A Review (2022, background): “that humans are involved in the decision-making process and that this level of involvement is based on the risk and the societal and environmental impact factor [183].”
Generalized Out-of-Distribution Detection: A Survey (2021, neutral): “A trustworthy visual recognition system should not only produce accurate predictions on known context, but also detect unknown examples and reject them (or hand them over to human users for safe handling) [1], [2], [3], [4], [5], [6], [7], [8].”
Survey of Explainable AI Techniques in Healthcare (2023, background): “To trust AI models, the European Union has proposed seven key requirements, including (1) human agency and oversight; (2) technical robustness and safety; (3) privacy and data governance, (4) transparency; (5) diversity, non-discrimination, and fairness; (6) social and environmental well-being; and ”
Should ChatGPT be Biased? Challenges and Risks of Bias in Large Language Models (2023, background): “By involving human experts in decision-making, we can ensure that AI models align with ethical principles and values, such as fairness, transparency, and accountability [125, 122, 138, 124, 107].”
Aligning AI With Shared Human Values (2020, background): “The demand for ethical machine learning (White House, 2016; European Commission, 2019) has already led researchers to propose various ethical principles for narrow applications.”
Croissant: A Metadata Format for ML-Ready Datasets (2024, neutral): “Croissant builds on these works by supporting the RAI documentation with a dedicated vocabulary extension [23].”