Trustworthy AI: Practices and challenges for enterprises adopting AI

Free Content Report March 2025 24 pages pdf

What you will learn

The Artificial Intelligence Act (AI Act) approved in 2024 by the European Union’s Legislative bodies, establishes a comprehensive regulatory framework to balance the benefits of Artificial Intelligence (AI) with the need to safeguard fundamental rights. This report explores the implications of the AI Act for European businesses, focusing on its risk-based approach, which categorizes AI systems into four levels of risk—unacceptable, high, limited, and minimal—and imposes varying requirements accordingly. Furthermore, this work provides actionable insights into how organizations can address challenges linked to the AI Act, leverage Trustworthy AI principles, and prepare for compliance with the regulation, ultimately enabling responsible AI adoption.

Table of Contents

1
The European legislative landscape: the AI Act
2
Implementing trustworthy AI in business projects
3
Towards trustworthy AI: where do large organizations stand?

Index of Graphs

1
Framework developed by the Observatory to map the requirements of the AI Act for businesses adopting AI - Politecnico di Milano

You will know how to answer these key questions

How can the European AI Act's regulatory requirements be integrated into the lifecycle of AI systems to ensure compliance and promote Trustworthy AI practices?
What practices have businesses implemented to ensure compliance with the AI Act, and what challenges do they face in adopting Trustworthy AI principles?

The methodology used in this Report

This analysis builds on the European regulatory framework established by the AI Act, examining its implications for businesses by mapping the regulation onto the value chain of AI systems according to the AI Project Value Chain model developed by the AI Observatory. To complement the regulatory analysis, the research included focus groups with 88 C-level executives and managers from large Italian enterprises. These interactive sessions explored business practices, challenges, and strategies for implementing Trustworthy AI, with greater emphasis on the principles of Fairness and Explainability.

We are here to provide information and assistance