AI is Becoming Regulated
Slovenia recently adopted a national legislative proposal to implement the European AI Act, the first comprehensive law regulating the use of artificial intelligence. The purpose of the Act is to establish clear rules of the game: how to develop and use AI in a way that delivers benefits while at the same time protecting human rights, safety, and user trust. For companies, this represents a challenge – but also an opportunity for digital solutions to become not only innovative, but also compliant with future regulation.
Why AI Needs Regulation
In recent years, the development of artificial intelligence has accelerated – from generative models like ChatGPT and similar, to advanced data analytics systems in healthcare, manufacturing, or logistics. Along with opportunities, however, come risks: biased algorithms, unexplained results, threats to privacy, or even manipulation of information.
The AI Act therefore introduces mechanisms that reduce risks and require technology providers to ensure safety, reliability, transparency, and compliance.
How the AI Act is Designed
The core logic of the Act is the classification of AI systems according to risk:
- Unacceptable risk – AI that could harm human rights (e.g., social scoring, behavioral manipulation) is prohibited.
- High risk – AI used in critical sectors (healthcare, transport, education, employment, security) is strictly regulated and must meet clear requirements regarding testing, documentation, and oversight.
- Limited risk – AI used for communication with users (e.g., chatbots) must clearly indicate that it is artificial intelligence.
- Minimal risk – AI that has no significant impact on people remains largely unregulated.
This approach ensures that legislation does not hinder innovation while at the same time protecting what is most valuable in society: safety and trust.
What This Means for Companies
For companies, the AI Act will mean:
- greater responsibility in the development and use of artificial intelligence,
- requirements for documentation and testing,
- the necessity of transparency and explainability of systems,
- inclusion of safety mechanisms already at the design stage (“security & safety by design”),
- the obligation to assess impacts on privacy and human rights in sensitive applications.
Safety First
One of the main messages of the Act is that AI is unacceptable without safety. This means:
- data protection,
- prevention of misuse,
- transparent algorithms,
- traceability of decisions.
In practice, this will mean that developers of artificial intelligence will have to build systems that are transparent and explainable, while also resilient to cyber threats.
The AI Act is not intended to limit innovation but to direct it toward responsible and safe practices. Companies that understand this will seize the opportunity: they will develop solutions that not only comply with the law but go beyond it, emphasizing ethics, reliability, and safety.
AI Act-Compliant Solutions
The company Kalmia is already developing solutions today that follow the key principles set out in the AI Act. Their approach is based on safety, reliability, and transparency.
AI Workshops and Consulting
- structured workshops and consultations to identify challenges and opportunities,
- helping companies design AI Act-compliant projects from the very beginning,
- developing a clear strategy for digital transformation.
Secure Digital Solutions
- safety mechanisms built in from the design and development phase,
- AI Act guidelines applied depending on the level of risk,
- ensuring compliance and user protection.
Reliable Architectures
- solutions based on modularity and proven methodologies,
- reducing risks and ensuring robustness,
- long-term stability and adaptability of systems.
Compliance Support
- a team of consultants with experience across industries,
- collaborating with companies to implement safe and transparent solutions,
- ensuring compliance with the new legislation.
Thus, Kalmia does not only offer technology, but also a reliable partner that helps companies understand and meet the requirements for the safe use of artificial intelligence – while at the same time building a competitive advantage with modern, trustworthy solutions.
By Borut Terpinc