Chip Revolution: What Drives the AI Industry

The rapid development of artificial intelligence (AI) is giving semiconductor stocks an impressive boost. Graphics cards, known as Graphics Processing Units (GPUs), are essential for generative AI, and companies like Nvidia, Advanced Micro Devices, and Taiwan Semiconductor are already considered early winners in this field. In the shadow of the GPU market lies the IT infrastructure, a sector that we believe is unjustly overlooked. GPUs find their place in data centers. It seems logical that with an increasing demand for GPUs, the demand for data center services would also need to rise. This was apparently the thought of billionaire investor Ken Griffin of Citadel Advisors. According to the fund's latest 13F filing, Citadel increased its stake in the data center stock Equinix by a remarkable 704% in the second quarter and now holds approximately 564,000 shares. Large Language Models (LLMs) like ChatGPT, Claude, and Gemini are currently extremely popular AI applications. With features ranging from image creation to software code generation to general search functions, these models are revolutionizing interaction in the workplace and on the internet. Although LLMs can respond to inquiries almost instantaneously, the underlying technology is far more complex. Generative AI, which performs tasks quickly, requires a prolonged development process. This is because machine learning applications undergo continuous training and inference tests. These processes drive the demand for data center resources. In Equinix's latest earnings call, CEO Adaire Fox-Martin drew an interesting comparison between the rise of AI and the cloud computing wave a decade ago. He noted that "AI training workloads are causing significant short-term demand" and that this also heavily impacts the demand for inference. Today, cloud services are a multi-billion-dollar business for tech giants like Amazon, Alphabet, Microsoft, and Oracle. One reason for this is the growing demand for digital infrastructure in parallel with the increasing interest among companies in data-driven decision-making.