Dataset Viewer
Auto-converted to Parquet Duplicate
month
stringdate
2023-01-01 00:00:00
2026-08-01 00:00:00
index_level
float64
84.4
345
2023-01-01
100
2023-02-01
87.49
2023-03-01
99.1
2023-04-01
84.4
2023-05-01
94.49
2023-06-01
87.1
2023-07-01
101.23
2023-08-01
93.43
2023-09-01
93.37
2023-10-01
95.39
2023-11-01
99.97
2023-12-01
92.16
2024-01-01
107.73
2024-02-01
107.62
2024-03-01
92.01
2024-04-01
94.84
2024-05-01
110.93
2024-06-01
121.9
2024-07-01
102.74
2024-08-01
113.72
2024-09-01
130.41
2024-10-01
115.48
2024-11-01
117.1
2024-12-01
132.03
2025-01-01
134.39
2025-02-01
150.28
2025-03-01
143.65
2025-04-01
147.43
2025-05-01
116.04
2025-06-01
153.21
2025-07-01
172.58
2025-08-01
196.88
2025-09-01
158.65
2025-10-01
187.58
2025-11-01
217.44
2025-12-01
165.09
2026-01-01
247.04
2026-02-01
209.79
2026-03-01
249.76
2026-04-01
258.46
2026-05-01
272.98
2026-06-01
306.87
2026-07-01
344.54
2026-08-01
335.36

Stockatomics AI Infrastructure Index

The Stockatomics AI Infrastructure Index tracks what it costs to build AI data centers, month by month. It brings together the benchmark prices of the GPUs, servers and memory that do the computing, the optics that connect them and the power and cooling equipment that keeps every site running, weighted by how much is spent on each. The index reads 100 in January 2023, the first month after the launch of ChatGPT set off the current wave of AI data center investment.

The live index, with its chart, the components behind it and how it is built, is at stockatomics.com/ai-infrastructure-index.

Files

ai_infrastructure_index.csv holds the monthly index readings from January 2023 to the latest published month.

Column Description
month The month of the reading, as the first day of the month
index_level The index level, with January 2023 = 100

ai_infrastructure_index_components.csv lists the components in the index and how much each one counts.

Column Description
component_id The component's Stockatomics identifier
component The component's name
sector The sector it belongs to on Stockatomics
weight_pct Its weight in the index, in per cent

How to read it

A reading of 200 means the components cost twice as much per unit as they did in January 2023, while a reading of 50 means they cost half as much. The index measures the price per unit of equipment, not the cost of computing power, so a rising index can sit alongside a falling cost per calculation.

How it is built

Every month, each component's Stockatomics benchmark price is compared with its level in January 2023, and the index combines those changes into a single figure. Each component has a fixed weight based on how much was spent on it in 2025, with no cap, so servers and GPUs carry the most weight. Components with a shorter price history join from their first month without changing earlier readings. The index is not adjusted for inflation and uses no averaging across months.

The benchmark price of every component, with its full history, trade flows, supply scores and the listed companies that make and buy it, is available in the Stockatomics Terminal.

Updates

The dataset is updated every month, when the latest benchmarks are published.

Licence and citation

Published under CC BY 4.0. Please cite as "Stockatomics AI Infrastructure Index", with the month of the reading and a link to https://www.stockatomics.com/ai-infrastructure-index.

@misc{stockatomics_ai_infrastructure_index,
  author       = {{Stockatomics}},
  title        = {Stockatomics AI Infrastructure Index},
  year         = {2026},
  howpublished = {\url{https://www.stockatomics.com/ai-infrastructure-index}},
  note         = {Monthly index, January 2023 = 100}
}

About Stockatomics

Stockatomics is a financial data platform for the physical economy. It builds monthly benchmark prices for critical minerals and the components behind AI data centers, power grids, electric vehicles, batteries, defence and robotics, and links each one to the listed companies that make and buy it.

Downloads last month
3