The key takeaway is that AI infrastructure is shifting from a GPU-supply story to an execution story. A headline capacity agreement matters, but delivered capacity, uptime, power availability, networking performance and customer utilisation will determine the eventual revenue and return on capital. I’m watching B: AI infrastructure most closely—especially power, cooling, high-speed networking and rack-scale integration. As clusters grow, a weak link in any of those areas can leave expensive GPUs underutilised. Compute leasing also looks promising, but contracts should be assessed by actual deployment schedules, cancellation rights, utilisation and renewal rates rather than their maximum headline value. For AI agents, inference demand may become the bigger long-term driver: repeated tool c