Artificial intelligence (AI) is now an integral part of our everyday lives. However, as AI becomes more widespread, there is a growing need for clear regulations to ensure its use is safe and responsible. The EU AI Act, which has been in force since August 2024, is the world's first comprehensive legal framework for AI. But what exactly does this mean for companies? And how can you prepare best for the new challenges? We clarify the most important questions.
What is the EU AI Act?
The AI Act is the European Union’s first comprehensive regulation governing the use of AI. The aim of the AI Act is to ensure that AI systems in Europe respect fundamental rights, safety and ethical principles. This applies in particular to high-risk applications, such as in the fields of medicine or cybersecurity. With the new law, the EU not only wants to get the risks of high-performance AI models under control, but also promote innovation and acceptance of trustworthy AI.
Why is it necessary?
The aim of the law is to create clear requirements and guidelines for the use of AI. It is intended as a protective measure to ensure that AI systems are not misused and violate human rights or ethical principles. At the same time, the legal framework is intended to minimize the administrative and financial burden, especially for small and medium-sized enterprises (SME).
What does the EU AI Act mean for companies?
By the end of 2025, companies must ensure that all employees who use and operate AI systems have specific AI skills. In addition, companies must check that they are not using any prohibited AI applications, such as social scoring or behavioral manipulation.
In addition to these general obligations, there is a classification of risk levels.
Different obligations apply depending on whether companies develop AI systems or use the AI systems of other companies. However, the majority of the obligations apply to high-risk AI systems (medical devices, cybersecurity, risk management, etc.).
In addition, a labeling obligation for AI-generated content will be introduced, which must be implemented by mid-2026. Providers of general purpose models, such as ChatGPT, must provide information on training data and technical documentation.
All specific information on the various risk levels and general information can be found here.
Concerns of companies
According to a recent study by Deloitte, 52.3% of companies fear that the AI Act will restrict their opportunities for innovation. Almost half (47.4%) of those surveyed even see the regulation as an obstacle to the introduction of AI applications in their own company.
Many companies fear overregulation and unclear regulations. However, the use of an “AI task force” provides companies with significant advantages. In this way, the rules can be implemented in good time and opportunities and developments for the future can be driven forward at the same time.
What does the AI Act mean for LinkThat?
At LinkThat, we have almost 10 years of experience in the development of AI solutions. From the very beginning, we have placed great emphasis on the quality and security of our AI systems. A dedicated team is responsible for training and labeling our AI. We take great care to use only high-quality data and not to include unclear or potentially problematic data in the training process in the first place.
Do we have any concerns about the AI Act at LinkThat?
“No, because we have always placed the highest demands on data protection and ethical principles. That’s why we don’t see the AI Act as a threat. It’s an opportunity to continue developing secure and innovative AI solutions.,” says Florian Jung, Head of AI Development.
Conclusion
The EU AI Act presents companies with new challenges, but it also offers great opportunities. In particular, companies that already focus on data protection, ethical principles and quality assurance have little to fear. The key lies in implementing the new regulations in good time. In the dynamic world of AI, the following applies more than ever. Those who act early will remain competitive in the future.