Law and Regulation

Compliance with legal requirements and standards is essential for the use of AI. With the new European AI Regulation of the EU (European AI Act), the European legislator has created guidelines for AI solutions that will apply directly in all EU member states. The legal framework is particularly relevant when companies want to develop their own AI systems. However, legal regulations also apply to the use of artificial intelligence.

Disclosure Obligation - What Companies Must Consider Now

The EU AI Regulation requires companies to train their employees in handling AI systems according to Article 4 of the EU AI Regulation since February 2, 2025. Find out who is affected, which measures are permissible, and why action is needed now.

Who is affected?

The regulation applies to every company or organization that uses AI tools. Operators of AI systems are defined in the AI Act as "natural or legal persons, authorities, institutions or other entities using an AI system under their own responsibility". This includes all professional applications - only private, non-professional uses are excluded.

Requirements for training or education

The AI Regulation does not define rigid requirements for training or education. Therefore, companies have the option to choose the type of training measures themselves. These can, for example, be carried out in the following formats:

• Online courses and webinars

• On-site training sessions and workshops

• Specialized literature and self-study

It is important that the training measures take into account the level of knowledge, experience, and specific application area of the employees concerned. The goal is to achieve a sufficient level of AI competence and thus ensure the responsible use of AI technologies.

Why is AI training or education important?

A thorough training or education not only ensures compliance with legal requirements, but also brings numerous benefits to companies:

• Risk minimization: Through trained employees, errors and potential for abuse can be reduced.

• Efficiency improvement: A better understanding of AI technologies enables a more effective use of the systems.

• Competitive advantage: Companies that use AI responsibly and competently can implement innovations more quickly.

European AI Act

Die europäische KI-Verordnung reguliert den Einsatz und die Entwicklung künstlicher Intelligenz innerhalb der EU. Sie zielt darauf ab, Risiken zu minimieren und die Sicherheit und Grundrechte der Bürgerinnen und Bürger zu schützen. Gleichzeitig soll sie Innovation und das Wachstum der KI-Technologie fördern, indem ein vertrauenswürdiger und rechtssicherer Rahmen für Entwicklerinnen, Entwickler, Anwenderinnen und Anwender geschaffen wird. Zu diesem Zweck werden KI-Systeme in vier Risikostufen (von unbedenklich bis inakzeptabel) eingeteilt. Systeme mit inakzeptablen Risiken sind verboten. Für hochrisikobehaftete KI-Anwendungen werden klare Vorgaben eingeführt. Diese Einteilung zielt darauf ab, ein Gleichgewicht zwischen der Förderung von Innovation und Technologieentwicklung und dem Schutz der Bürgerinnen und Bürger vor potenziellen Missbräuchen oder Schäden durch KI-Technologien zu schaffen. Sie wurde im März 2024 durch das europäische Parlament verabschiedet und muss größtenteils binnen zwei Jahren befolgt werden.

Legal advice

For companies that want to build their own AI systems, the legal environment is particularly important. Less complex challenges exist for companies that use off-the-shelf solutions, but a careful examination is also necessary here. A well-founded legal advice is sometimes essential to navigate this rapidly changing field and minimize legal risks.

Regulation of specific industries

In some industries such as healthcare, financial services, or transportation, specific regulatory requirements may apply that concern the use of AI technologies.

Data protection and copyright

The protection of data and intellectual property are further important aspects for safeguarding the rights of individuals or organizations. The legal and regulatory framework for AI continues to evolve dynamically. As technology often advances faster than the corresponding laws and standards, companies find themselves in a space where clear legal guidelines are still in development. This requires a proactive approach and regular review of the relevant legal situation to ensure compliance and minimize legal risks.

Protection against discrimination

Manufacturers of AI systems must consider product liability laws that hold them responsible for damages caused by errors in the system. For example, in the healthcare sector, AI systems are used where a malfunction can have far-reaching consequences for a patient. Another sensitive area related to AI is financial services, where, for instance, a flawed training dataset can lead to significant financial loss. Appropriate regulation must ensure what requirements are to be placed on the design and monitoring of such systems.

By integrating legal and ethical aspects into your technology strategy, you can not only minimize legal risks but also strengthen your reputation and credibility in the public eye.

You can find more information on the AI Compass topic "Law and Regulation" in our download portal.