Powering AI, Paying the Price? Data Centres, Household Costs and Local Environmental Changes

Overview

Artificial intelligence depends on large and rapidly expanding physical infrastructure. Data centres provide the computing power needed to train and run AI systems, cloud services and data-intensive applications, but they also require substantial electricity, water, land and network capacity. As governments and firms invest heavily in AI infrastructure, an important economic question is whether the benefits and costs of this expansion are shared evenly across households, businesses and local communities.

This project investigates whether the location, density and growth of data centres are associated with changes in household energy prices, water prices and water availability, as well as broader local environmental conditions. Depending on data availability, the research will also examine outcomes such as air quality, land use change, vegetation and biomass indicators, agricultural productivity, renewable energy investment and local economic activity. The central aim is to develop large-scale empirical evidence on the local and regional consequences of AI infrastructure growth.

The project combines applied economics, environmental economics, spatial data science, energy and water policy, artificial intelligence and geospatial analysis. The research team has already assembled a global database of approximately 6,299 data centre locations from 2025. The VRES project will refresh, clean and extend this database using available public sources; standardise facility names, geographic coordinates, operator information and available capacity or opening date details; and identify overlapping or duplicate records across sources.

The resulting geocoded dataset will be linked to publicly available national, subnational and local indicators. Where feasible, the student will help construct spatial exposure measures based on proximity to, or concentration of, data centres and estimate associations using fixed effects, event study, difference-in-differences or related spatial econometric approaches. The analysis will be carefully scoped to the available data, with an initial focus on a defined set of countries, states, regions or environmental outcomes rather than attempting to estimate every possible impact globally within the VRES period.

This is a timely national and international research issue. Governments are seeking to expand domestic compute capacity and attract investment while ensuring that infrastructure growth remains affordable, reliable and environmentally sustainable. The project will contribute evidence relevant to energy policy, water governance, regional development, environmental planning and responsible AI infrastructure.

Research engagement

The student will:

• Conduct a review of existing literature on data centre growth, electricity demand, water use, environmental externalities, household affordability and spatial economic impacts;
• Refresh, clean and document the existing global data centre location database using publicly available sources and reproducible Stata, Python or R workflows;
• Develop methods to identify likely duplicate facilities across multiple data sources and construct a transparent, analysis-ready geocoded dataset;
• Link data centre locations to selected energy, water, environmental, land use, agricultural or socioeconomic outcome datasets at the most detailed geographic level available;
• Construct spatial measures of local exposure to data centre development, such as distance-based buffers, regional facility counts or capacity proxies;
• Conduct descriptive, mapping and econometric analysis using methods such as panel data regression, fixed effects, event studies, difference-in-differences or spatial analysis where appropriate;
• Document the findings, interpret results cautiously and identify priorities for a larger cross-country and subnational research program.

Research activities

The student will work with Dr Steve Bickley and other members of the behavioural economics/AI research team at the ARC BITA Centre, in collaboration with researchers from the Department of Economics at the University of Oviedo. They will gain practical experience in applied economics, Python, R or Stata, data engineering, geospatial data analysis, reproducible research workflows, regression modelling, data visualisation, research documentation and academic writing.

• The project aims to produce a draft manuscript that includes the following:
  • Introduction: Overview of AI infrastructure growth, policy relevance and study significance.
  • Literature Review: Review of research on data centres, energy, water, environmental externalities and local economic impacts.
  • Research Aims/Objectives and Questions/Hypotheses: Clear articulation of the empirical questions and expected relationships.
  • Data and Methodology: Description of data-centre database construction, outcome-data linkage, spatial measures and empirical strategy.
  • Results and Discussion: Presentation and interpretation of descriptive, spatial and econometric findings.
  • Conclusions and Future Work: Summary of insights, limitations and recommendations for future research and policy.
  • Reference List/Bibliography
  • Appendices: Data dictionary, data linkage/cleaning protocol, robustness checks and supplementary results where appropriate.
• Generate a brief 2–3 slide presentation to present the research at the Faculty of Business and Law VRES Showcase at the conclusion of the program.
• (Optional) The student may be eligible to present the findings to an audience of academics and industry partners at the annual BITA conference in February/March 2027, subject to project progress and the student’s interest.
• (Optional) The longer-term aim is to develop the work into a co-authored academic paper submission. However, the primary deliverable of this VRES project specifically is a final draft manuscript and reproducible analysis package. In other words, no work beyond the VRES period is required and any further work on the paper beyond this period would be entirely voluntary and optional for the student.

Research skills

They will gain practical experience in applied economics, Python, R or Stata, data engineering, geospatial data analysis, reproducible research workflows, regression modelling, data visualisation, research documentation and academic writing.

Outcomes

• The project aims to produce a draft manuscript that includes the following:
  • Introduction: Overview of AI infrastructure growth, policy relevance and study significance.
  • Literature Review: Review of research on data centres, energy, water, environmental externalities and local economic impacts.
  • Research Aims/Objectives and Questions/Hypotheses: Clear articulation of the empirical questions and expected relationships.
  • Data and Methodology: Description of data-centre database construction, outcome-data linkage, spatial measures and empirical strategy.
  • Results and Discussion: Presentation and interpretation of descriptive, spatial and econometric findings.
  • Conclusions and Future Work: Summary of insights, limitations and recommendations for future research and policy.
  • Reference List/Bibliography
  • Appendices: Data dictionary, data linkage/cleaning protocol, robustness checks and supplementary results where appropriate.
• Generate a brief 2–3 slide presentation to present the research at the Faculty of Business and Law VRES Showcase at the conclusion of the program.
• (Optional) The student may be eligible to present the findings to an audience of academics and industry partners at the annual BITA conference in February/March 2027, subject to project progress and the student’s interest.
• (Optional) The longer-term aim is to develop the work into a co-authored academic paper submission. However, the primary deliverable of this VRES project specifically is a final draft manuscript and reproducible analysis package. In other words, no work beyond the VRES period is required and any further work on the paper beyond this period would be entirely voluntary and optional for the student.

Skills and experience

• An interest in applied economics, environmental economics, energy policy, water policy, geography, data science, sustainability, AI or regional development;
• Some proficiency in data entry, data analysis and statistical techniques;
• Experience using software such as Stata, R or Python would be beneficial;
• Experience with GIS, geospatial data, mapping, APIs, web data extraction, SQL or geospatial Python/R libraries is desirable but not required; and
• Interest in careful, transparent and reproducible empirical research.

Start date

2 November, 2026

End date

19 February, 2027

Location

QUT Gardens Point campus (Z Block, Level 7), with some work able to be completed online/remotely subject to mutual agreement between the student and their VRES supervisor.

Additional information

The project provides an existing 2025 global data centre location database, python scripts from the initial web harvest and related data documentation, along with regular supervision. The student will receive guidance on reproducible coding, data source documentation, spatial data matching/linkage and applied econometric analysis.

Before commencement, the student will read a short set of papers on the economics of data centres, energy and water demand, environmental externalities, spatial analysis and AI infrastructure policy. The VRES placement is not expected to verify every data centre globally; instead, the student will produce a transparent refreshed database and a carefully defined analysis-ready sample suitable for the initial empirical study.

Keywords

• Applied economics
• Artificial intelligence
• Energy prices
• Electricity demand
• Water availability
• Environmental economics
• Geospatial data analysis
• Sustainability

Contact

Dr Steve Bickley

+61 7 3138 5115

s.bickley@qut.edu.au