Outlook 2027: AI expenditure flows heavily into infrastructure

The outlook for 2027 shows that the majority of AI investment and spending continues to flow into infrastructure, rather than directly into the models themselves. This pattern is influenced by the increasing complexity and scalability of AI systems. The infrastructure includes not only the physical servers and storage, but also the software and cloud platforms required to deploy, operate and secure AI models. Investment in infrastructure is critical to ensuring the performance, efficiency and security of AI systems, especially at a time when AI is increasingly involved in critical areas such as health, finance and industry.

Another aspect of AI development is the increasing globalization of data and technologies. Sources report that US companies are likely to sell AI training data and reinforcement learning environments to Chinese AI companies. This raises questions about data protection compliance and digital sovereignty. Dissemination of training data across borders can bring both opportunities and risks, as it can accelerate the development of AI models, but could also jeopardize sensitive data and national interests.

The safety and efficiency of AI agents is another key issue. The Cloud Native Computing Foundation (CNCF) supports projects aimed at making AI agents more secure and efficient, especially in the cloud and container environment. These efforts are particularly relevant as AI agents are increasingly used in companies and organizations to automate processes and support decisions. Ensuring the trustworthiness and transparency of AI systems is therefore a central focus of developments.

Another highlight of AI development is the continuation of research projects using AI in biology and medicine. For example, Google has developed a new DNA model called AlphaGenome 2, which takes into account the effects of genetic changes in different cell types. This model could significantly accelerate the search for disease causes and the development of new therapies. Such advances show how AI is able to analyse complex biological systems and thus provide a crucial impetus for medical research.

## What this means for users The self-hosted, GDPR-compliant infrastructure of xynap enables a secure and efficient integration of AI systems that meets the requirements of modern technologies.