Artificial Intelligence: The Government launches the first national regulatory system: schools, work, justice, and security are at the heart of the reform.

The Council of Ministers has given preliminary approval to the implementing decrees of the AI ​​law: mandatory training, protections for workers and citizens, and rules for the police and judiciary. Investments of up to €1 billion are planned for startups and innovation.

The Government is accelerating its efforts on Artificial Intelligence, making Italy one of the first European countries to adopt a comprehensive national regulatory framework to govern the use of new technologies. The Council of Ministers, meeting on June 10, 2026, at Palazzo Chigi, chaired by Prime Minister Giorgia Meloni, approved in preliminary examination two important legislative decrees implementing Law No. 132 of 2025 on Artificial Intelligence.

The reform aims to combine technological innovation, protection of fundamental rights, economic competitiveness, and national security, fully consistent with the European AI Act, the EU regulation governing the use of artificial intelligence in the European Union.

Italy among the first European countries with a comprehensive regulation on AI

According to the Government, the primary objective is to ensure the development of Artificial Intelligence based on an "anthropocentric" approach, in which technology represents a tool at the service of humanity and not a substitute for human responsibilities.

The philosophy of the regulation is based on the principle that AI can support decisions, services, training, and production activities, but cannot replace human judgment or undermine fundamental human rights.

The Government emphasizes that the new legislation was developed in constant dialogue with the European Commission and represents a national integration of the AI ​​Act, without creating regulatory overlaps with the EU framework.

Artificial Intelligence in Schools: New Programs and €100 Million to Fight Digital Addiction

One of the most significant interventions concerns the school system.

Artificial Intelligence will become a permanent part of Italian education through the updating of secondary school curricula and the integration of advanced technologies and generative AI into study programs.

Among the new features expected:

• incorporating AI into civic education;
• enhancing STEAM skills;
• strengthening of school guidance;
• ongoing teacher training;
• creation of territorial technical-ethical committees to support schools.

Particular attention is paid to the phenomenon of digital addiction. The decree allocates €100 million for teacher training aimed at preventing risks arising from social media, digital platforms, and artificial intelligence tools used by minors.

Families will also be involved in promoting digital well-being and combating algorithmic conditioning and technological dependence.

Universities, research, and ITS Academy: AI enters all training programs

The reform also involves universities, AFAM institutions and ITS Academies.

The new courses will have to integrate dedicated content:

• to the functioning of AI systems;
• to the interpretation of the outputs;
• to the legal profiles;
• to cybersecurity;
• to the impacts on fundamental rights.

The goal is to develop interdisciplinary skills that enable understanding both the technological and regulatory and ethical aspects of Artificial Intelligence.

A central role will be entrusted to ANVUR, which will monitor the quality of the educational offering and be able to guide future academic incentive policies.

Public Administration: Mandatory training for employees

The reform introduces a comprehensive training program for public administration personnel.

Public employees will be involved in the following courses:

• basic digital literacy;
• professional retraining;
• advanced training for managers and digital transition leaders.

According to the Government, AI could help simplify administrative procedures, improve citizen services, and speed up decision-making times, while always maintaining human control over final decisions.

Healthcare: AI Enters Continuing Medical Education

For the first time, Artificial Intelligence will formally enter the ECM (Continuing Medical Education) program.

Healthcare professionals will need to acquire specific skills not only in the technical use of the tools, but also in ethical, deontological, and legal aspects.

Clinical responsibility will always remain with the doctor.

The training will also target healthcare managers to improve service organization, waitlist management, and waste reduction in the National Health Service.

Work: Layoffs are prohibited, and decisions are made solely by algorithms.

One of the most relevant passages concerns the world of work.

The decree establishes that:

• hiring;
• layoffs;
• disciplinary sanctions;
• changes in the employment relationship

they cannot be decided exclusively by automated systems.

The final decision must be made by a natural person with decision-making powers.

Workers will also have the right to obtain an understandable explanation of decisions that affect them and to know what role AI systems, if any, play in the decision-making process.

The provision regarding dismissals is particularly severe: those adopted in violation of the ban on automation will be considered null and void.

Magistrates and Justice: AI Will Not Replace Judges

In the judicial sector, the Government reaffirms the principle of the centrality of the magistrate.

Artificial Intelligence can be used as a support tool for:

• jurisprudential research;
• office organization;
• document management.

However, it cannot replace human judgment.

The training of magistrates will be entrusted to the Higher School of the Judiciary and will focus on the technical, regulatory, and organizational aspects of AI.

Facial recognition and police activity: judicial authorization and strict limits

One of the most sensitive issues concerns the use of Artificial Intelligence by law enforcement.

The decree introduces stringent rules for biometric and facial recognition.

Real-time biometric identification will be permitted exclusively for:

• prevent serious threats to public safety;
• search for missing persons;
• identify victims of kidnapping or trafficking.

Authorization from the judicial authority will be required and the maximum duration will be limited to 15 days, unless justified extensions are granted.

Furthermore, the creation of biometric databases through massive and indiscriminate collections of images from the web is prohibited.

The Government clarifies that the system does not introduce any form of generalized surveillance of the population.

New protections for those harmed by AI systems

The decree also strengthens civil protection for people harmed by Artificial Intelligence systems.

Among the measures envisaged:

• access to the system's technical documentation;
• presumption of causal link;
• possibility of taking action in the court closest to one's residence;
• direct action against insurance companies.

The goal is to reduce the evidentiary difficulties that often characterize disputes related to particularly complex technologies.

New crime for those who tamper with the security of high-risk AI systems

The reform also introduces the new Article 437-bis of the Criminal Code.

The law will punish:

• failure to adopt security measures;
• high-risk AI system disruption;

when such conduct determines a concrete danger for:

• people's lives;
• public safety;
• state security.

Not only natural persons, but also the companies involved may be held liable under the provisions of Legislative Decree 231/2001 governing the administrative liability of entities.

One billion euros for startups and AI ecosystem development

Alongside regulation, the Government is focusing on industrial growth.

The law provides for the allocation of up to one billion euros from the Venture Capital Support Fund to the development of the national Artificial Intelligence ecosystem.

The data released in the press release highlight that:

• the Italian AI market reached 1,8 billion euros in 2025;
• annual growth was 50%;
• over 300 million euros have already been invested;
• more than 150 startups have been supported;
• over 1.000 highly qualified workers are employed in the financed companies.

The SophIA hub, dedicated to Artificial Intelligence and cybersecurity, will also be operational in 2026, with approximately €30 million earmarked for technology transfer and the development of deep-tech startups.