AI’s power and water consumption is worrying the agriculture sector: ‘Don’t forget that it is also required for us to grow food’

Rapid global buildout of AI data centers is creating acute demand for electricity and water that could compete with agriculture and pressurize local energy markets—Singapore briefly paused new centers in 2019 and U.S. states like Virginia are seeing rising power prices—prompting warnings that data-center energy use may crowd out human food production. Panelists at the Fortune Innovation Forum said rising populations and protein demand will require big productivity gains; vertical-farming startup Agroz claims up to 500% higher yields using 20x less water, while experts urged adoption of practical measures today (greenhouses, drip irrigation, better livestock monitoring, fermentation) to raise output and cut emissions (rice farming is cited as responsible for ~8% of global emissions); the discussion highlights investment and policy implications across energy grids, water infrastructure and scalable agtech and food-innovation plays focused on efficiency and quality.

Analysis

Global data-center expansion tied to the AI buildout is creating a direct competing demand for electricity and water that threatens agricultural inputs, as highlighted by Gerard Lim’s comment that data-center energy use could “leave the human sectors out.” Policymakers have already reacted: Singapore paused data-center investments in 2019 over resource concerns and U.S. states such as Virginia are experiencing rising electricity prices in areas of heavy data-center construction, indicating localized stress on grids and cost pass-through risks.

Panelists framed the supply-side agricultural challenge against rising demand: speakers cited an expected ~50% increase in food demand by 2050 and shifting diets toward more protein, which raises pressure on yields today. Technology offers mitigation: vertical-farming startup Agroz claims up to 500% higher yields while using 20x less water, showing the scale of efficiency gains possible through controlled-environment agriculture.

Speakers urged pragmatic adoption of existing productivity levers—greenhouses, drip irrigation, fermentation, and better livestock monitoring—rather than only state-of-the-art solutions; Richard Skinner pointed to these as deployable today. Rice cultivation’s ~8% share of global emissions underscores the emissions-reduction upside from irrigation and practice changes, and the confluence of energy, water and food security creates cross-sector investment and regulatory risk that investors should monitor.

Key Decisions for Investors

• Prioritize exposure to scalable agtech and controlled-environment agriculture businesses that can demonstrate material water and yield improvements (for example, technologies claiming multi-hundred percent yield gains with 20x water savings),
• Monitor regional utility and regulatory risk around major data-center hubs (e.g., Virginia and jurisdictions that have paused data-center growth) and reduce concentration in equities or REITs with high exposure to those grids,
• Allocate tactically to companies offering low-tech, high-impact agricultural upgrades (greenhouses, drip irrigation, livestock monitoring) because these solutions are deployable now and can materially reduce water use and emissions,
• Consider thematic positions in water infrastructure and sustainable food/fermentation platforms as hedges against agricultural supply constraints and to capture demand from higher-protein diets,
• Watch for policy shifts or moratoria on data-center permitting and local electricity-price signals as near-term catalysts that could re-rate both data-center and agriculture-related equities