As the artificial intelligence sector experiences rapid expansion, top executives from across the supply chain are warning that the industry is hitting significant, tangible constraints. Experts from firms including ASML, Google Cloud, Applied Intuition, Perplexity, and Logical Intelligence recently gathered to discuss the reality behind the hype, highlighting that hardware, energy, and physical data requirements are creating a bottleneck that could persist for years. Christophe Fouquet of ASML noted that despite a massive surge in chip manufacturing, the sector will likely remain supply-limited for the next several years, meaning major tech giants may struggle to acquire the high-end processors they require to scale their operations.

Energy consumption has emerged as a parallel crisis, with Google Cloud exploring unconventional solutions, including the potential for data centers in orbit to access more abundant power, despite the significant engineering hurdles involved. Beyond hardware, there is a growing debate over whether the current reliance on scaling large language models is the most efficient path forward. Researchers are increasingly investigating ‘energy-based models’ that aim to replicate human-like reasoning rather than simple pattern recognition, potentially offering a more sustainable and faster alternative for robotics and physical AI applications.

Furthermore, the integration of AI into physical infrastructure—such as autonomous defense vehicles, mining equipment, and agricultural robotics—is introducing new layers of geopolitical complexity. Governments are becoming increasingly protective of their borders, showing resistance to foreign-controlled physical AI systems. Industry leaders emphasize that as AI moves from purely digital search tools to autonomous physical agents, the focus must shift toward granular security, sovereign control, and addressing labor shortages in critical fields like agriculture and logistics where human workers are becoming increasingly scarce.