Artificial Intelligence (AI) is transforming every sector, from healthcare to finance. However, as algorithms take center stage in decision-making, the need to ensure that these technologies are safe, ethical, and understandable is also growing.
This is precisely the focus of the Blue AI podcast episode titled “AI Regulation and Governance”, where Lidia, the host, speaks with Teresa Jambrina, Director of Corporate Legal and AI Governance at NTT DATA.
The episode addresses essential topics such as the role of law in artificial intelligence, European regulation (the AI Act), legal risks in business, and the importance of transparency and trust in AI development.
At DeepXAI, we found this episode particularly relevant because it reflects a core idea that we also champion in our course, XAI Applied to Deep Learning.
The Role of Law in Artificial Intelligence
During the conversation, Teresa Jambrina highlights a key point: the law cannot fall behind technological progress.
AI introduces new ethical and legal dilemmas, such as accountability for automated decisions or model transparency, that require new regulatory frameworks.
Therefore, the role of law is not just to set limits, but to provide legal certainty to companies and users alike.
Regulation helps define responsibilities, protect rights, and prevent the misuse of technologies that, by their very nature, can directly impact people's lives.
In this sense, the law becomes an ally for the responsible development of AI. The key lies in finding a balance between innovation and protection: allowing technology to advance, but within well-defined ethical and legal boundaries.
The European Framework: The AI Act
One of the central themes of the podcast is the European regulation of artificial intelligence, known as the AI Act. This regulation, driven by the European Union, seeks to establish a common framework for the safe, ethical, and transparent use of AI across member states.
The AI Act classifies AI systems based on their risk level:
- Unacceptable risk:Systems that pose a threat to fundamental rights (for example, cognitive manipulation or mass surveillance). These will be prohibited. prohibidos.
- High risk:Applications in sensitive areas such as healthcare, education, or employment. They must comply with strict requirements regarding transparency, documentation, and human oversight. requisitos estrictos de transparencia, documentación y supervisión humana.
- Limited or minimal risk:: sistemas que podrán usarse libremente, aunque se fomentará la autorregulación ética.
This risk-based approach marks a turning point in AI governance.
For the first time, a global institution is establishing clear criteria to evaluate the trustworthiness and safety of intelligent systems. Even more importantly, the regulation champions a fundamental principle that also guides our work at DeepXAI: explainability.
AI systems will need to be understandable and auditable, ensuring that both developers and users can comprehend why an algorithm makes a specific decision.
Governance and Legal Risks in Business
Another highlight of the episode is AI governance within organizations.
Having advanced models is not enough; companies need to establish control structures, internal policies, and supervision protocols to ensure the responsible use of technology.
Teresa Jambrina mentions that the primary legal risks for companies stem from three areas:
- Lack of transparency: When algorithmic decisions cannot be explained.
- Data bias: Which can lead to discrimination or unfair outcomes.
- Absence of human oversight: Completely delegating decisions to a machine without critical human review.
To mitigate these risks, companies must adopt active AI governance that includes:
- Regular model audits.
- Multidisciplinary teams (legal, technical, and ethical).
- Clear documentation of the algorithm's lifecycle.
- Internal training on ethics and regulations.
Governance is not just a bureaucratic formality; it is a competitive advantage. Companies that develop trustworthy AI aligned with regulations will be the ones that earn the trust of the market and their users.
Transparency and Trust: Pillars of Responsible AI
One of the greatest challenges of current AI is the opacity of its models. Deep learning systems, in particular, can act as "black boxes", they generate accurate results, but those results are difficult to explain.
This is where Explainable AI (XAI) comes in, a field that seeks to make the internal reasoning of models visible, allowing humans and machines to work together.
Transparency not only improves technical understanding but also strengthens trust. If a user or regulator can understand why a system made a decision, it becomes much easier to accept, validate, and improve it.
At DeepXAI, we believe that trust cannot be imposed; it must be built. And it is built by combining three pillars:
- Ethics: Using data that respects human and social values.
- Regulation: Complying with current laws and regulatory frameworks.
- Explainability: Ensuring that AI is understandable, traceable, and verifiable.
Toward an Ethical and Compliant Artificial Intelligence
The conversation between Lidia and Teresa Jambrina on Blue AI highlights that the future of AI depends not only on technical breakthroughs but also on its ability to integrate into society ethically and legally.
European regulation is a major step forward, but its success will depend on collaboration among tech experts, legal professionals, and business leaders.
AI governance cannot be left to a single sector; it requires a cross-disciplinary vision that combines technical knowledge, legal criteria, and ethical sensitivity.
In this context, education is key. Professionals who understand the relationship between explainability, ethics, and regulation will be better equipped to lead AI projects that are transparent, trustworthy, and sustainable.
That is why at DeepXAI we promote educational initiatives like our XAI Applied to Deep Learning course, where we teach how to build explainable models that align with the principles of trust and responsibility promoted by the European Union.
Innovation, Yes But with Responsibility
The debate over AI regulation is not about slowing down innovation, but about giving it direction and purpose.
An artificial intelligence that is ethical, explainable, and subject to legal governance is not an obstacle to progress; it is the very foundation of its legitimacy.
Only then can technology contribute to human development without violating fundamental rights.
The Blue AI episode reminds us that the future of AI will only be as positive as the responsibility with which we build it.
That future begins today, with every ethical decision, every explainable model, and every company that commits to transparency.
You can watch the full video here: