The Tax Agency (AEAT) manages one of the most powerful and sensitive databases in Europe. In an era where the algorithm threatens to become judge and party, the institution marks an unbreakable ethical red line in the face of the rise of Artificial Intelligence. In the challenges section of the Orbital Vision 2026 yearbook, José Borja Tomé, Director of the Tax IT Department, details how to balance technological efficiency with taxpayer guarantees.
When talking about digital transformation in Spain, the Tax Agency is usually the elephant in the room: the most advanced body, the one that crosses the most data and for many the most feared. However, given the rise of Generative Artificial Intelligence and predictive models, a legitimate concern arises among citizens: Are we moving towards a scenario where a “black box” automatically decides, for example, who receives a sanction or who is inspected?
In an extensive interview for the section Challenges of Orbital VisionJosé Borja Tomé does not hide and does not choose to deny the advantages that AI can provide. In this way, it dispels possible ethical and privacy doubts by defining the Supervised Artificial Intelligence that governs the institution. “The automated administrative actions dictated by the Tax Agency will not, in any case, rest exclusively on the result obtained from an AI system,” says the manager.
This declaration is much more than a technical standard, it is a declaration of principles. It means that although the AEAT systems are capable of processing millions of declarations and detecting fraud patterns invisible to the human eye in milliseconds, the final decision (the administrative act that has legal effects on the citizen) will always maintain human supervision. The algorithm proposes, alerts and assists, but it is the official who validates and decides.
From inspection and sanction to preventive assistance
Beyond the ethics of the algorithm, the interview collected in the yearbook reveals a fundamental strategic shift in the mission of the AEAT. Traditionally, technology was used to “hunt” fraud in retrospect. Today, the objective is to use that computing power to prevent the error from occurring.
“The real challenge in the transformation of the AEAT is to take advantage of the available capabilities to ensure easier and fairer tax compliance,” explains Borja Tomé in the text. The vision of the future is not an administration that sanctions more, but one that bothers less.
This translates into a model of “compliance by design”. Thanks to AI, the Agency can offer increasingly complete drafts (not only of personal income tax, but also of VAT and Corporations), virtual assistants that answer complex questions in natural language and early warning systems that notify the taxpayer of possible discrepancies before they file their return. Technology thus becomes a tool of equity: by simplifying bureaucracy for the honest citizen, the Agency frees up high-value human resources to focus on the fight against complex fraud and the underground economy.
A mirror for the private sector
In Orbital Vision 2026it also explores how the AEAT is leading the definition of governance frameworks that could serve as a standard for the private sector. It is not enough for the technology to work; It must be explainable, robust and non-discriminatory.
The conversation with José Borja Tomé delves into how these systems are internally audited and how the administration prepares for an imminent future where interaction with the State will, inevitably, be digital by default. For any manager, whether in the public or private sphere, understanding how the organization with the most data in Spain manages the risk and ethics of AI is an essential management lesson.
If you want to read the full interview and much more, download Orbital Vision 2026 for free https://orbital-vision-2026.getresponsewebsite.com/
