On September 17, 2025, the Italian Senate definitively approved the new law on artificial intelligence (Bill no. 1146-B “Provisions and Delegation to the Government on Artificial Intelligence”).
This law is intended to regulate the development and use of AI through a legal framework that establishes principles, competent authorities, critical sectors, investments, and safeguards. The goal is to foster innovation and the competitiveness of organizations using AI, while at the same time protecting fundamental rights.
But what are the key points of the Artificial Intelligence Law, along with its advantages and limitations? And how will it impact businesses, public administrations, and citizens? Here’s a detailed breakdown.
The 5 Key Points of the Artificial Intelligence Law
The new Italian law on artificial intelligence includes several provisions covering different aspects.
Here are the 5 main points:
- General principles: AI must always remain “human-centered,” ensuring transparency, safety, respect for human dignity, and protection of personal data.
- Sensitive sectors: The law identifies certain areas where AI use will be particularly regulated—healthcare, public administration, justice, labor, education, and even sports. In these fields, AI can be a useful support tool, but final decisions must remain in human hands.
- Privacy and child protection: The law strengthens alignment with the GDPR, requires clear consent for children under 14, and promotes the use of anonymized or synthetic data, especially in health research.
- Governance: Oversight will be handled by the National Cybersecurity Agency and the Agency for Digital Italy. Additionally, a National AI Strategy will be updated every two years. The law also explicitly declares alignment with the European AI Act.
- Economic investment: €1 billion will be allocated to support startups, SMEs, and research projects—a strong signal of trust in those investing in the sector.
Pros and Cons of the Artificial Intelligence Law
Like any major reform, this law comes with both strengths and challenges.
Strengths include regulatory clarity—businesses and institutions now finally have clear rules—and concrete support for innovation, which could help Italy’s tech ecosystem make a significant leap forward. Equally important is the focus on protecting rights, placing Italy among the countries most attentive to the ethical use of AI.
Challenges, however, remain. Many crucial details—such as civil liability and technical criteria—will only be defined in implementing decrees. Until then, some uncertainty will persist. Moreover, overly strict rules could risk slowing AI adoption, especially for small businesses with fewer resources to dedicate to compliance.
Practical Consequences: What Changes for Businesses, Public Administrations, and Citizens
The key question is: what does this law mean in practice? The answer varies depending on the stakeholder:
- Businesses
- Public administrations
- Citizens
Consequences for Businesses (especially startups and SMEs)
Companies will need to adapt their AI development processes to meet requirements for transparency, traceability, and accountability. This may involve investments in audits, documentation, algorithm quality control, ethical data use, and privacy by design. While businesses in the sector could benefit from the allocated funds, they must prepare now to comply with criteria that will be set in implementing decrees—possibly relying on specialized advisors such as Italy & Partners.
Consequences for Public Administrations (PA)
Public administrations will also face changes. They will be able to use AI systems to improve efficiency, but ultimate decision-making must remain human.
This will require staff training, transparency toward citizens, guaranteed access to information, and likely a review of internal processes to integrate AI governance.
Consequences for Citizens
For citizens, the new law means stronger protection of their rights. They will gain clearer information about AI use, the right to oppose their personal data being used for AI training, better protection for minors, and more control over who is responsible for automated decisions.
However, the complexity of these technologies may make it difficult to fully understand what happens “behind the scenes” of an algorithm. Therefore, the effectiveness of safeguards will depend heavily on how well the explanatory and implementing rules are drafted.
What Will Be the Concrete Impact of the Law?
In the short term (within 12–18 months of its enactment), we will see the adoption of delegated decrees defining technical criteria, sanctions, the classification of high-risk systems, and responsibilities. Companies will need to closely monitor these developments to avoid penalties and seize opportunities.
In the medium to long term, if well implemented, the law could consolidate a trustworthy and competitive AI ecosystem, attract investments, build public trust, and improve the quality of public services.
For these reasons, the Italian Artificial Intelligence Law does not mark the end of the journey, but the beginning. Once the decrees are issued in the coming months, companies, public administrations, and citizens will fully understand the extent of the change.
If properly applied, this law could make Italy not only safer and more ethical in its use of AI but also more competitive on an international scale. The challenge will be to balance innovation and protection—a difficult but increasingly necessary equilibrium.