The debate on artificial intelligence ethics is evolving, shifting from an approach based on abstract principles to the concrete need for transparency and control over AI systems, to ensure decisions are fair and understandable.
What happened
The discussion around ethical AI has long been dominated by the definition of guidelines and values. However, a recent Wired.it article highlights how this perspective is insufficient. The author argues that ethics does not reside intrinsically within AI, but in the human choices that design and implement it. The real challenge, therefore, is not to teach AI to be ethical, but to build systems that are inherently transparent, allowing humans to understand their decision-making processes and intervene to ensure fairness and accountability. This paradigm shift underscores the need for tools and mechanisms that enable "acting within" AI, rather than simply observing it passively. The goal is to move from imposed morality to the capacity for moral action through design.
Why it matters
This vision has profound implications for individuals, businesses, and society. Without transparency, artificial intelligence systems can perpetuate or amplify existing biases, erode public trust, and lead to unfair decisions in critical sectors such as justice, finance, or healthcare. A lack of control means users and organizations cannot correct errors or unintended deviations, making accountability difficult to assign. For businesses, adopting opaque AI carries significant reputational and legal risks, as well as hindering responsible innovation. For workers, understanding how AI influences decision-making processes becomes crucial for reskilling and adapting to new professional landscapes. Society as a whole risks delegating fundamental decisions to incomprehensible systems, compromising the principles of democracy and social justice. The ability to intervene and modify AI behaviors is fundamental to preserving human autonomy.
The HDAI perspective
The Human Driven AI perspective aligns perfectly with this evolution of the debate. We firmly believe that artificial intelligence must be a tool at the service of humanity, designed with the primary goal of enhancing human capabilities, not opaquely replacing them. True ethical AI is not an intrinsic attribute of the machine, but the result of robust governance and transparent design that places human beings at the center. This means moving from "post-hoc" ethics to "ethics by design," where transparency, interpretability, and the possibility of human intervention are integrated from the earliest stages of development. This approach will be central to discussions at the upcoming HDAI Summit 2026, where we will explore how to translate these principles into concrete practices and effective regulations to ensure AI always operates for the benefit of society. The key is to build AI systems that are not only fair but also empower humans to verify and guide their fairness at every stage of their lifecycle.
What to watch
Regulatory evolution, such as the EU AI Act, is already pushing towards greater transparency and accountability, classifying AI systems by risk and imposing specific obligations. It will be crucial to observe how these regulations are implemented and what technical standards will be developed to make transparency operational. Focus will increasingly shift towards auditing and certification tools that can ensure the real-world compliance and trustworthiness of AI systems, providing decision-makers with the necessary tools for effective control and proactive risk management.