Home Corporate Communication News Artificial Intelligence: DDL 1146, AI Act and ISO/IEC 42001

Artificial Intelligence: DDL 1146, AI Act and ISO/IEC 42001

Three Converging Perspectives on Responsible Artificial Intelligence

Artificial Intelligence: DDL 1146, AI Act and ISO/IEC 42001
Artificial Intelligence: DDL 1146, AI Act and ISO/IEC 42001 In the increasingly complex framework of Artificial Intelligence regulation, a coherent regulatory and operational triangle is emerging today between three fundamental instruments: the Italian Bill 1146, the European AI Act Regulation, and the international standard ISO/IEC 42001:2023.

Each of them approaches the topic of AI from a different, but profoundly complementary, perspective.

DL 1146 – The value and institutional foundation

Bill No. 1146 proposes a comprehensive intervention that, while still in progress, marks an important step in building public governance of AI in Italy. It:
  • establishes general principles that the use of AI must adhere to : human-centeredness, transparency, non-discrimination, privacy protection, human supervision, inclusion;
  • provides for delegation to the Government to adopt legislative and regulatory measures in strategic sectors (justice, healthcare, labor, public administration);
  • Strengthens the role of AgID and ACN as public bodies responsible for promoting and overseeing the responsible use of AI.
In this sense, the DDL acts as an ethical-legal guide , offering a value-based orientation on which to base technical standards and management systems.

AI Act – The binding European regulatory framework

Regulation (EU) 2024/1684, known as the AI Act, represents the first global attempt to regulate Artificial Intelligence through horizontal and binding legislation.

The AI Act:
  • adopts a risk-based approach , classifying AI systems into 4 levels: unacceptable, high, limited, minimal;
  • imposes specific obligations for high-risk systems , including: risk management, data governance, transparency, traceability, human oversight and post-market surveillance;
  • provides for significant penalties for failure to comply with the requirements (up to 35 million euros or 7% of annual turnover).
The regulation defines in detail what must be done to ensure safety, reliability and respect for fundamental rights.

ISO/IEC 42001 – The operational framework for compliance and excellence

ISO/IEC 42001:2023 , the first of its kind to define an AI Management System (AIMS), is the practical tool that enables organizations to:
  • transform the principles of DDL 1146 into policies, roles, operating procedures and measurable indicators;
  • implement all AI Act requirements in a structured and scalable manner, with a focus on: risk assessment, oversight, transparency, technical documentation, and continuous improvement;
  • map and control the entire AI lifecycle , through 38 controls organized into 11 control objectives (Annex A).

In this context, ISO/IEC 42001 provides the “how” to do it , enabling native integration with other ISO standards ( ISO/IEC 27001 , ISO/IEC 23894, ISO/IEC TR 24027), in an integrated management logic.

_______________________________________________________________________________________

CSQA TRAINING OFFER

ON DEMAND COURSE: AI ACT – ARTIFICIAL INTELLIGENCE LITERACY

ONLINE COURSE: LEAD AUDITOR ISO/IEC 42001:2023 - ARTIFICIAL INTELLIGENCE

Would you like to have more informations?

Contact us

Newsletter subscription form

You need information, contact us

One of our staff will answer or contact you as soon as possible

Fields marked with an asterisk (*) are required