As the dialogue around data centers and their impact on the energy industry becomes increasingly pervasive, the rapid expansion of data centers also raises concerns about grid capacity.
But what are the actual economic drivers behind data center expansion, and what do we know about the size of the market for data center capacity?
Software now has marginal costs, largely from the energy needed for model training, inference, and deployment. Like other goods, computational power comes with both marginal cost and value—the cost is easy to estimate but the value is less obvious. Eventually, the value of generative AI will equal its cost, creating market equilibrium. The central question remains: how many new data centers will this drive and how real is the demand?
In this CRA Insights authors Maxime Cohen and Christopher Russo address the following key considerations in order to better understand AI’s impact on the energy sector:
- What’s driving the growth of generative AI?
- How do the economics of data centers compare with the economics of energy?
- How will market dynamics be impacted by computational capacity
- What may future growth in the sector look like
Read the Insights here.
