The rapid expansion of artificial intelligence (AI) technologies is significantly affecting electricity use in the global data center sector. According to Goldman Sachs Research, AI is projected to drive a 165% increase in data center power demand by 2030, reaching 122 gigawatts (GW) of data center capacity online by the end of the decade. This growth is fueled by the continued increase in data, compute, and connectivity from digitalization and cloud migration, as well as the scaling of new technologies like AI.
Currently, data centers consume around 55 GW of power globally, with AI accounting for 14% of this demand. However, AI's share is expected to rise to 27% by 2027. The surge in power demand will require significant investment in power infrastructure, with Goldman Sachs estimating around $720 billion in grid spending through 2030.
The increasing power demand from data centers raises concerns about sustainability and carbon emissions. To mitigate the environmental impact, data center operators are exploring alternatives like nuclear energy, renewable energy, and advanced cooling systems. For instance, nuclear power could provide a consistent and reliable source of energy for data centers, aligning with the stated ambitions of most data center operators to reduce their carbon footprint.
