Europe and Italy Redefine AI: The New Rules for Professionals

Artificial intelligence has left the regulatory grey zone. The European AI Act and Italy's Law No. 132/2025 now define what professionals can automate, and what must remain human judgment.

Arash Namjoo Fard

11/4/20254 min read

A sleek navy and black-themed blog card displaying an article about AI in professional services, with a clean white title overlay.

Artificial intelligence has stopped being a futuristic idea. It's now part of our everyday work, in firms, offices, and even the smallest professional practices. Accountants, lawyers, consultants, engineers, all of us are starting to rely on intelligent systems to process information, draft documents, and make decisions faster than ever before.

Until recently, this transformation was happening in a regulatory grey zone. There were tools, there was enthusiasm, but not much structure. That era is over. With the entry into force of the European AI Act (Regulation 2024/1689) and Italy's Law No. 132/2025, Europe has finally drawn the lines. For the first time, we have a system that doesn't just allow AI, (it encourages it) but within limits that protect what truly matters: human judgment and professional responsibility.

A European Vision: Regulating by Risk

The AI Act, which took effect in August 2024, is the world's first comprehensive legal framework for artificial intelligence. It takes a pragmatic view: not all AI systems carry the same risks, so they shouldn't all face the same controls.

Some systems, like social scoring or subliminal manipulation, are now completely banned. Others, those that affect access to justice, employment, or public services, are classed as high-risk and must meet strict standards for transparency, documentation, and human oversight. And then there are the everyday systems, chatbots, invoice readers, data classifiers, that make our work easier without touching fundamental rights. These come with lighter rules, as long as users are informed and privacy is respected.

The rollout is gradual: the ban on unacceptable systems began in February 2025; new obligations for generative AI start in August 2025; and full compliance for high-risk systems will follow in August 2026. Europe isn't trying to slow innovation, it's building a framework of trust so that innovation can last.

Italy's Step Forward

Italy took that European foundation and built on it. On October 10, 2025, Law No. 132/2025 came into force, making Italy the first EU country to adopt its own national law on AI.

Unlike the AI Act, it doesn't deal with every industry. Instead, it focuses where technology meets people most directly: the professions, healthcare, labour, public administration, and justice. For professionals, the message is clear. Article 13 of the law says that AI can only be used for instrumental and support activities, and that the prevalence of intellectual work must always remain with the human professional. In practice, this means AI can prepare, compare, and analyse, but it cannot decide or interpret on your behalf. Judgment remains the beating heart of professional work.

The law also introduces a new and deeply ethical principle: transparency toward clients. Anyone using AI must explain, in simple language, what systems are being used, for what purpose, and how human supervision is ensured. A short disclaimer at the end of an engagement letter is no longer enough, clients deserve to understand, clearly, how technology plays a role in the service they're paying for. This isn't about bureaucracy; it's about trust. And in our professions, trust is everything.

The Human Dimension of a Digital Revolution

At the centre of both the European and Italian approaches lies the same conviction: AI is here to serve people, not replace them. Italy's law gives that principle legal weight. It's a safeguard for clients, but also for professionals. In a world where automation tempts us to delegate more and more, the law reminds us that our greatest value isn't speed or efficiency, it's understanding. Professional responsibility can't be outsourced to a machine. The ability to interpret, to doubt, to question, these remain the exclusive domain of human intelligence.

That's why Italy's model is often described as anthropocentric, human-centred by design. It's not nostalgia for a pre-digital world. It's a way to make sure that progress doesn't erase the very qualities that make professional work meaningful.

A New Kind of Professionalism

To put these principles into practice, Italy also created two new institutions: the National Cybersecurity Agency (ACN), which supervises and monitors AI systems, and the Agency for Digital Italy (AgID), which promotes innovation and manages certifications. This dual structure captures the spirit of the entire reform: innovation supported by control, progress guided by responsibility.

For accountants, lawyers, and consultants, this means moving from informal experimentation to structured adoption. Using AI will now require documentation, internal policies, and staff training, not to create red tape, but to demonstrate competence and accountability. And that's the real turning point. Yes, AI will automate many parts of our work, from data entry and reconciliation to drafting and analysis. But the final decision, the interpretation, and the responsibility will always be ours. The smarter our tools become, the more our judgment matters. Firms that embrace this balance, using AI to save time and reinvesting that time in higher-value work, will lead the way. Those who see it only as a compliance burden will fall behind.

A Compass, Not a Brake

The combination of the AI Act and Italy's Law No. 132/2025 doesn't just regulate technology, it redefines professionalism. It asks us to be informed, transparent, and accountable in how we use digital tools. Seen this way, the law isn't a constraint; it's an invitation, to innovate consciously, to integrate technology without losing our intellectual independence, and to turn compliance into credibility. Artificial intelligence will keep evolving. But the essence of our professions, judgment, ethics, and trust, must remain human. That's the real meaning of this reform: not a brake on progress, but a compass to navigate it.