What is artificial intelligence regulation and why is it crucial for businesses and governments
Artificial Intelligence is evolving almost daily, but that does not make it an experimental technology; rather, it is a technology that is already part of our lives and has enormous transformative potential. As an expert in AI regulation, my view is that it plays, and will play an even more decisive role in, the way we work, how we interact with one another, between businesses and consumers, and between citizens and public authorities; it will influence the decisions we make, the decisions that affect us made by third parties, and access to public services. It will affect the very concept of citizenship and the society in which we live.
A technology with this potential and impact requires compliance with a set of rules, at least in those areas that affect us so significantly. That is why we must ensure that this technology is developed and used in a safe, transparent manner that respects people’s rights. As AI gains the ability to learn, generate content, recommend actions or influence decisions, the need to establish these rules also increases.
Good governance of artificial intelligence is key to building trust, which is an essential concept here. In the digital economy, it is not merely a reputational indicator: it is an economic and social asset. When people perceive that an organisation protects their data, manages cybersecurity risks appropriately and uses AI responsibly, their willingness to adopt digital services, use them more intensively and maintain more stable relationships with companies increases.
For this reason, regulating AI should not be seen merely as a compliance obligation, but as a means of creating the conditions for a society that trusts the digital world and, from there, generates tangible socio-economic value that boosts digital adoption and innovation.
How is the regulation of artificial intelligence evolving globally and what impact does it have on businesses
AI is evolving very rapidly, and every step it takes paves the way for the next. These successive technological developments in AI are occurring in rapid succession but at an exponential rate, and we will soon see AI applied to multi-agent ecosystems and AI-native business processes, where oversight and control will come under even greater strain. These will need to take new forms, based not so much on documentary compliance but on a technical approach to operational compliance. Regulation must keep pace with these new forms of supervision and control, where traceability and transparency will play a decisive role.
The transformative potential of AI takes on a geostrategic character that is far more evident than in previous technologies. Hence, different approaches are emerging depending on the region.
Europe has opted for a model that regulates market access for products and services on a risk-based basis. The European AI Regulation classifies systems according to their potential impact: prohibited uses, high-risk systems, transparency obligations and systems for general use. It is an approach that seeks to protect the safety of AI-based products and services, guarantee fundamental rights and, at the same time, provide stable legal rules.
In other regions, such as the United States, the approach has traditionally been more sector-specific, combining guidelines, standards, self-regulation and specific rules, with parallel regulation by the states, as is already the case with data protection. In Asia, too, we see diverse models, ranging from strategies heavily focused on innovation to highly centralised and directed regulatory frameworks.
For businesses, this evolution has a direct impact. It is no longer enough to develop or acquire AI solutions from a purely technological perspective. It is necessary to assess risks, establish controls, train teams, vet suppliers and ensure that systems comply with requirements and risk mitigation measures.
But there is an important point: regulation and responsible technology should not be viewed merely as mechanisms to avoid penalties, incidents or reputational damage. They must also be understood as an active lever for value creation. An organisation that builds trust can improve the uptake of its services, reduce friction, increase customer retention and foster stronger relationships with users, employees, partners and public authorities.
AI thus becomes part of companies’ strategic agenda not only because of its technological potential, but also because of its economic, social and competitive impact.
How AI regulation affects technological development and adoption in companies such as Telefónica
In a company such as Telefónica, AI regulation has a significant impact because artificial intelligence is applied across multiple areas: networks, customer service, process automation, data analysis, cybersecurity, energy efficiency, digital services and solutions for businesses and public administrations.
Regulation requires the incorporation, from the outset, of criteria relating to accountability, security, privacy, transparency, explainability, traceability and human oversight. This means that compliance cannot be added at the end of the process, but must be integrated from the design stage and throughout the system’s entire lifecycle.
Telefónica has had Artificial Intelligence Principles in place since 2018, updated in 2024, and a governance model approved in 2023 and applicable across the entire company. This model enables the potential of AI to be harnessed whilst protecting people and minimising risks through assessments and tools designed to reduce bias, improve explainability, reinforce fairness in decision-making and take environmental efficiency into account.
This approach ties in directly with Telefónica’s role as the best gateway for citizens to digital technologies: facilitating the use of the most innovative technologies, with the highest quality of service and a commitment to protecting users.
Regulation should therefore not be viewed solely as a restriction. Applied correctly, it can become a competitive advantage. And in a sector such as telecommunications, where trust is inseparable from service quality, this aspect is particularly important.
Key challenges in regulating artificial intelligence to balance innovation and technological growth
The main challenge is to strike the right balance. Insufficient regulation can lead to risks, abuses, legal uncertainty or a loss of trust. But excessively rigid regulation can slow down innovation, especially in a technology that evolves much faster than traditional regulatory cycles.
That is why regulation must be proportionate, technologically neutral and based on the specific use of AI, not on the technology in the abstract. Applications posing unacceptable risk must be banned; high-risk applications must be subject to enhanced obligations; and low-risk uses must be able to develop within more flexible frameworks, with transparency and best practice.
Telefónica has advocated a vision of better regulation; this does not always mean less regulation, but rather better regulation.
- evidence-based, not based on intuition
- proportionate, tailored to the actual risk
- consistent with the entire regulatory system, not based on isolated rules
- providing legal certainty and predictability
- based on aggregate impact assessments and ex-post reviews
- not creating regulatory asymmetries
- future-proof, not reactive and at the mercy of technological developments as they occur
Nor should we forget that the pace of technological innovation often outstrips the pace of regulatory development. That is why self-regulation, when applied seriously and responsibly, can complement formal regulation and help put the principles of responsible technology into practice.
We must also change the way we assess regulation and responsible technology. Traditionally, its impact has been measured from a reactive perspective: avoiding penalties, reducing incidents, preventing litigation or minimising reputational damage. All of this is necessary, but not sufficient. Responsible technology also actively generates value. In this sense, responsible technology should be seen as a lever for growth and not merely as a compliance cost.
How should companies prepare to comply with AI regulation and ensure responsible use
Companies must prepare with a holistic approach. AI regulation does not affect only legal or technical teams. It involves business, technology, privacy, security, procurement, human resources, sustainability, compliance, communication and customer relations.
The first step is to gain visibility: knowing which AI systems are used, where, for what purpose, with what data, with which suppliers and at what level of risk. Without an inventory, governance is impossible. This inventory need not be based on self-declaration but on automated monitoring and supervision mechanisms
The second step is to define clear responsibilities. AI requires governance models that establish who approves, who supervises, who manages risks, who reviews suppliers and who is accountable for potential impacts.
The third step is to incorporate controls throughout the entire lifecycle: design, development, procurement, deployment, use and monitoring. At Telefónica, the principles of responsible AI include AI that is people-centred, transparent and explainable, fair and inclusive, respectful of privacy and security, committed to the environment, and based on responsibility and accountability throughout the value chain.
But as well as complying, companies must learn to measure the value of doing things right. This involves developing indicators that link responsible technology practices to concrete results: reduced incidents, lower fraud, fewer complaints, greater acceptance of AI-based features, lower drop-off rates in digital processes, improved perception of control, security and transparency, and greater willingness to use digital services.
Responsible AI is not just about corporate compliance but primarily about corporate culture. It requires companies that design, purchase, sell, integrate or use AI systems to understand both their opportunities and their limitations.
The role of artificial intelligence in telecommunications and its regulation in strategic sectors
Artificial intelligence is playing an increasingly significant role in telecommunications. It can help optimise networks, anticipate incidents, improve service quality, strengthen cybersecurity, personalise customer service, automate operations and advance energy efficiency.
Telecommunications constitute an essential infrastructure for the digital economy. Networks underpin businesses, public administrations, public services, personal communications, the connected industry, smart cities, education, digital health and critical digital services.
That is why the telecommunications sector must combine innovation with security, resilience and responsibility. It is not enough for a solution to be technically efficient; it must be reliable, traceable, robust and respectful of users’ rights.
At Telefónica, this ties in with a fundamental idea: having the best network to access the most innovative technology and offer more and better services to customers.
The quality of these services will increasingly be measured by customer trust.
Regulation must allow the potential of AI to be harnessed in the management of digital infrastructure, particularly when it comes to low-risk applications, without imposing disproportionate burdens that hinder innovation. At the same time, it must require appropriate safeguards where AI may affect rights, security or essential services.
Furthermore, there is a social dimension. A business ecosystem that uses technology responsibly helps ensure that more people can benefit from innovation.
How does Telefónica contribute to responsible artificial intelligence that is aligned with regulatory frameworks in Europe?
Telefónica has been working for years on a vision of responsible artificial intelligence, centred on people and aligned with human rights, privacy, security, sustainability and accountability.
This vision is based on a simple idea: AI should enhance what it means to be human, not replace human responsibility. AI systems can improve processes, expand capabilities and generate new opportunities, but they must always be at the service of people.
Telefónica’s Principles of Artificial Intelligence establish a code of conduct for trustworthy, human-centred AI. These principles are aligned with proposals from international organisations such as the OECD, the Council of Europe and UNESCO, and are implemented through an internal governance model approved in 2023.
In practice, this means working towards AI that is transparent and explainable, avoiding incomprehensible black boxes; AI that is fair and inclusive, reducing bias and discrimination; AI that respects privacy and security, based on privacy and security by design; AI committed to the environment; and AI that is responsible and accountable throughout the value chain.
This approach also reinforces Telefónica’s institutional role. The company aspires to be a benchmark in Europe, not only as a telecommunications company, but as a strategic player in digital transformation, technological innovation, digital sovereignty and the regulatory debate.
Contributing to responsible AI therefore means more than simply complying with regulation. It means actively participating in the construction of a framework that allows for innovation with confidence, protects people and strengthens European technological leadership.
What impact does the regulation of artificial intelligence have on user trust and the adoption of digital solutions?
Trust is one of the decisive factors in the adoption of any technology. In the case of artificial intelligence, this is even more so.
Users want to know when they are interacting with AI, how their data is used, whether an automated decision can be reviewed, what level of human oversight exists, what security safeguards are in place, and what mechanisms they have to lodge a complaint or seek an explanation.
Regulation helps to create a common framework of safeguards. It requires transparency, accountability, risk assessment, oversight and, in certain cases, enhanced controls. All of this helps users feel more protected.
But trust is not built on regulations alone. It also depends on how companies implement responsible technology in their systems, how they manage risks, how they respond to errors, and how they remain consistent with the principle that technology is at the service of people.
At Telefónica, this idea ties directly into our ambition to offer more and better services to customers. Trust is not an incidental element of the digital experience; it is a central part of service quality.
Furthermore, digital trust has tangible economic effects: it can increase the willingness to adopt digital services, improve repeat usage, reduce churn, facilitate the uptake of innovative services, strengthen brand advocacy and reduce friction in customer relationships.
In short, trustworthy AI can accelerate digitalisation. Opaque or poorly governed AI can slow it down. That is why regulation, governance and corporate responsibility are complementary elements.
What trends in artificial intelligence regulation will shape the coming years in Europe and globally?
The first will be the practical application of regulations already approved. The challenge will shift from designing regulatory frameworks to implementing them within organisations: AI inventories, risk assessments, controls, audits, supplier governance and internal training.
The second will be the development of technical standards, guidelines and codes of good practice. Regulation requires specific tools to ensure it can be applied in a consistent, verifiable and proportionate manner.
The third will be the growing focus on generative AI, general-purpose models and agentive systems with greater autonomy. These systems pose specific challenges regarding transparency, intellectual property, security, traceability, accountability and human oversight.
The fourth will be international convergence. AI is global, and its regulation cannot evolve in a completely fragmented manner. Telefónica has advocated for the need to strengthen international cooperation to establish common principles and avoid regulatory fragmentation, with the aim of protecting rights and fostering innovative solutions.
And the fifth will be the need to better measure the socio-economic value of responsible technology. Until now, there has been much talk of risks avoided, but it will become increasingly important to also quantify the active benefits: trust, adoption, reduced friction, social well-being, equal access to technology, greater competitiveness and a greater capacity to generate value for customers and users.
Future regulation must not only specify what cannot be done with AI. It must also help create the conditions for responsible AI to generate trust, innovation, growth and social progress.
Conclusion
The regulation of artificial intelligence should not be seen as a barrier, but as a prerequisite for innovation to be sustainable, reliable and socially valuable.
The real challenge is not to choose between regulation or innovation, but to build responsible innovation, capable of generating economic and social value without compromising the protection of rights, security, privacy and trust.
For Telefónica, this vision is fully in line with its role as the primary gateway for citizens to digital technologies: a company that facilitates the use of the most innovative technologies, with the highest quality of service and a commitment to protecting users.
In strategic sectors such as telecommunications, AI can help us build smarter networks, more efficient services, more personalised digital experiences and solutions with a positive impact on society. But for all that potential to be realised, we need AI that is well-governed, transparent, explainable, secure, inclusive and aligned with people.
Governance of artificial intelligence is, ultimately, a way of steering technological progress towards a more human future. And that future will depend not only on the power of technology, but on our ability to turn trust into a genuine economic and social asset.
