The AI Act and the Literacy Revolution: Two Years Later, Taking Stock

Two years after its entry into force, the European AI Act is proving to be much more than a simple regulatory framework. What many initially considered yet another bureaucratic hurdle is revealing itself to be a catalyst for cultural change in companies. And I, who have been following the evolution of artificial intelligence in the European business landscape for years, must admit that this is one of the most positive surprises in recent times.

When Mandatory Training Becomes Opportunity

I still remember the reactions when the AI literacy obligation was announced. Skeptical entrepreneurs, managers worried about costs, HR managers wondering how to fit yet another mandatory course into already overloaded calendars. Yet, what is happening in Italian and European companies is something unexpected: mandatory training is transforming into a genuine moment of collective awareness.

We're not talking about simple technical or compliance courses. We're talking about an AI literacy process that is changing the way people perceive artificial intelligence. From an obscure and incomprehensible threat to a tool that can be understood, governed, and used responsibly.

The Paradigm Shift in SMEs

Small and medium-sized enterprises, traditionally more skeptical of complex technological innovations, are experiencing an interesting transformation. Mandatory training has forced them to confront AI, but this confrontation has opened unexpected doors.

I've seen manufacturing companies discover artificial intelligence applications for optimizing production processes. Professional firms that have started questioning how AI could support, not replace, their employees' work. Small retail businesses that have understood the risks of automated customer scoring systems, consciously deciding how to use them.

This is true literacy: not knowing how to program an algorithm, but understanding what it means to integrate AI into your business, what questions to ask, what risks to consider.

The AI Act as a Compass, Not a Chain

When the AI Act was approved, the debate focused almost exclusively on constraints. Risk classification system, obligations for providers, bans on applications considered dangerous. All right, all necessary. But what emerges today is a different aspect: the AI Act is functioning as a compass to navigate complexity.

Risk Classification as a Thinking Tool

The distinction between high-risk, limited-risk, and minimal-risk systems is not just a bureaucratic exercise. It has become a mental framework that helps companies think critically about their technological choices.

When a company asks "is this personnel selection system high-risk?", it's not just verifying a regulatory requirement. It's reflecting on the impact that technology will have on people's lives. It's considering bias, transparency, accountability. It's, ultimately, doing what any responsible organization should do before implementing powerful technology.

Transparency as Competitive Value

Another positive side effect concerns transparency. Companies that have embraced the AI Act's disclosure obligations are discovering that openly communicating how they use artificial intelligence is not an admission of weakness, but a competitive advantage.

Consumers, increasingly aware, appreciate knowing when they interact with an automated system. Employees trust company decisions more when they understand AI's role in processes that affect them. Business partners positively value the technological maturity demonstrated by those who know how to govern these tools.

The Missing Awareness

If there's one result that the AI Act and mandatory training are achieving, it's this: bringing awareness. And awareness is the prerequisite for every responsible choice.

Beyond Hype and Technophobia

For years, the debate on artificial intelligence has oscillated between two extremes. On one hand, the uncritical enthusiasm of those who saw AI as the solution to every problem. On the other, the irrational fear of those who imagined apocalyptic scenarios of total human replacement.

Mandatory literacy is creating an intermediate space, that of critical thinking. People are learning to distinguish between real capabilities and exaggerated promises. To recognize where AI can truly make a difference and where it's just marketing.

This capacity for discernment is perhaps the most precious gift that European regulation is giving us. Not technical experts who know everything, but citizens and professionals who know how to ask the right questions.

The Human Factor Returns to Center Stage

Paradoxically, the greater the awareness about artificial intelligence, the more the importance of the human factor emerges. Companies that have seriously tackled mandatory training are discovering that AI doesn't eliminate the need for human skills, but makes them even more crucial.

We need people who can interpret algorithm outputs, who understand model limitations, who maintain final responsibility for decisions. We need, ultimately, human intelligence to govern artificial intelligence.

Light and Shadow: Not Everything Is Perfect

It would be naive to paint an entirely positive picture. The road to AI literacy is still long and presents significant obstacles.

The Uneven Quality of Training

Not all AI literacy programs are equal. Excellent programs exist that tackle the subject with depth and pedagogical skill. But superficial courses also exist, created more to satisfy the formal obligation than to transfer real knowledge.

The risk is that a disparity is created: companies and professionals who have access to quality training develop real awareness, while others limit themselves to obtaining a certificate without truly understanding the subject. A collective effort is needed to raise the quality bar.

The Gap Between Large and Small Organizations

Large companies have resources to implement structured training programs, hire internal experts, build solid governance. Small businesses often struggle to find time and budget to go beyond the bare minimum.

Here Europe could do more: targeted incentives, public high-quality training platforms, accessible consulting support. AI literacy cannot become a privilege of the most structured organizations.

The Temptation of Formal Compliance

There's always the risk that companies focus on formal compliance rather than substance. Filling out documents, filing certificates, creating procedures on paper, without this translating into real cultural change.

Against this temptation, continuous vigilance is needed, from authorities but also from professionals themselves. The AI Act works if it's experienced as an opportunity for improvement, not as a nuisance to be disposed of as quickly as possible.

Looking to the Future

Two years is a short period to evaluate the impact of complex regulation like the AI Act. Yet the signs are encouraging. AI literacy is taking hold, awareness is growing, companies are beginning to integrate ethical and risk considerations into their decision-making processes.

The International Domino Effect

The European approach is influencing other markets. Countries and regions that were initially skeptical are evaluating similar frameworks. Multinationals operating in Europe are extending AI governance standards to extra-European operations as well, because they've understood that managing artificial intelligence responsibly is simply good business.

AI literacy thus becomes a global competitive advantage for European companies, which develop distinctive competencies in the conscious management of technological innovation.

The Next Frontier: From Awareness to Action

The challenge of the coming years will be transforming acquired awareness into concrete action. Knowing what AI is or what risks it entails is not enough. We need to develop operational capabilities: multidisciplinary teams that know how to design responsible systems, effective audit processes, correction mechanisms when something goes wrong.

Literacy is the starting point, not the finish line. But it's an indispensable starting point, which Europe had the courage to impose when other markets left everything to the free market.

A Bet That's Paying Off

Looking back to two years ago, when the AI Act was just a text approved by the European Parliament and the training obligation raised perplexity, I can say that the European bet is paying off. Not perfectly, not without obstacles, but it's paying off.

Companies are more aware. Professionals have tools to navigate complexity. Public debate on artificial intelligence is more mature, less ideological, more anchored to reality.

Mandatory AI literacy works because it has forced everyone, from CEOs to operational employees, to stop and reflect on what it really means to integrate artificial intelligence into our professional and personal lives. It has created a common language, a shared framework, a starting point for more responsible choices.

Europe, often criticized for its regulatory approach, has demonstrated that regulation can be an instrument of progress when well-conceived and accompanied by investments in training. The AI Act is not perfect, but after two years we can say it's a good law. A law that, instead of merely prohibiting or permitting, has chosen the more difficult but more fruitful path: educating.

And in a world increasingly shaped by artificial intelligence, education is the most precious resource we have.