What are the semiconductor supply chain risks for AI infrastructure, and which countries are most exposed?

The semiconductor supply chain for AI infrastructure faces significant risks from surging demand outpacing supply, leading to chip shortages, swollen supplier backlogs, and escalating prices for memory (RAM) and storage components [1][2][10][11]. These bottlenecks, driven by AI's high compute needs for data centers scaling to gigawatt levels, constrain build times and limit global capacity, with additional pressures from concentration in chip design and manufacturing that heighten competition risks [5]. Geopolitical factors amplify vulnerabilities, including US export controls on advanced AI chips to China and potential disruptions to key chipmaking materials from the Iran crisis [3][6][9][12]. Countries most exposed include China, facing suspended exports of Nvidia and AMD AI chips and requirements for foreign investment pledges to access US semiconductors [6][12]; the UK, uniquely vulnerable to the datacenter investment boom's risks amid broader AI infrastructure scarcity [7]; and South Korea, concerned about Iran-related disruptions to semiconductor materials essential for its chip industry [9].