Photo: images.theconversation.com
· rights & takedowns The Conversation article by Michael Vardon reports Australia lacks granular, public data on water and electricity consumption by new AI data centres. The piece cites Sam Altman saying Australia could become a "data centre capital of the world" and notes that Australian industries used 17.6 million megalitres of water in 202324, with agriculture accounting for about two-thirds, per national statistics reported in the article. The Conversation highlights a sectoral value contrast: a megalitre used in data centres is valued at $2.3 million compared with $4,600 per megalitre in agriculture. The article flags energy as the larger environmental concern, and it cites research reported in the piece suggesting electricity demand from data centres could outstrip available renewable generation and prompt new gas-fired capacity. The author calls for more granular measurement of both water and power use for policy and planning. What happened The Conversation article by Michael Vardon reports that Australia currently lacks granular public data on the water and electricity consumption of newly proposed and built AI data centres . The article notes OpenAI chief executive Sam Altman said Australia could become a "data centre capital of the world," and it cites national statistics showing Australian industries consumed 17.6 million megalitres of water in 202324, with agriculture using nearly two-thirds, as reported in the piece. The article contrasts estimated sectoral value: a megalitre in data centres at $2.3 million versus $4,600 per megalitre in agriculture, per figures presented in the article. The Conversation also cites research suggesting electricity demand from data centres could outstrip clean power from renewables and lead to new gas plants. Editorial analysis - technical context Industry-pattern observations: measuring environmental impact of compute-heavy facilities requires device-level and site-level telemetry, cooling-system water-accounting, and energy-source disclosures. Companies and grid operators in comparable markets have used PUE and metered coolant water flows to build forecasts, but those metrics are rarely public at fine granularity. Industry context For practitioners: the article frames energy rather than water as the primary systemic risk, because large-scale compute can change peak demand profiles and renewable integration needs. Public-value-per-megalitre comparisons in the article show data centres are economically dense water users, but the figure does not alone resolve trade-offs with agricultural or ecosystem water needs. What to watch For practitioners and policymakers: track the emergence of site-level disclosures, regulatory requirements for water and energy reporting, and grid-supply studies that quantify incremental peak and annual demand from AI clusters. The Conversation calls for more granular measurement before major planning or policy decisions are finalised. Scoring Rationale This story matters to practitioners because emerging AI infrastructure can materially shift grid demand and resource planning; however, it is primarily a data-availability and policy-question piece rather than a new technical breakthrough. Practice interview problems based on real data 1,500+ SQL & Python problems across 15 industry datasets — the exact type of data you work with. Try 250 free problems