EU AI Act and local hardware trends reshape AI landscape

The landscape of artificial intelligence is undergoing significant regulatory and structural shifts. Under the EU AI Act, new categories for mid-sized companies have been introduced to provide a transition zone between SMEs and large corporations, easing technical documentation requirements. Regulatory bodies like BaFin are also increasing oversight of AI systems used in financial services to ensure compliance with risk management and data protection laws.

In the private sector, a trend toward local AI processing is emerging. Companies are increasingly investing in dedicated hardware, such as Apple’s M-series chips or specialized Nvidia systems, to run large language models locally. This approach aims to reduce reliance on expensive cloud subscriptions, lower latency, and enhance data security by keeping sensitive information on-site.

However, the integration of AI brings challenges regarding responsibility and public perception. The BVK warns that while AI can provide information, it cannot assume legal responsibility for insurance or financial decisions. Similarly, research from the University of Graz indicates a gap between the EU’s legal risk classifications and how the public perceives AI dangers. Experts emphasize that human accountability remains essential, as AI systems require continuous monitoring and rigorous testing to prevent errors in regulated environments.