Energy Resilience Meets AI

AI infrastructure and energy resilience are becoming a joint sovereign-risk equation that will increasingly differentiate countries’ growth, inflation, currencies, and interest rates.

• The International Energy Agency expects global data-center electricity demand to more than double by 2030 to about 945 terawatt-hours, while US data-center consumption is projected to exceed the energy used by aluminum, steel, cement, chemicals, and other energy-intensive goods combined.
• The authors argue that the Iran war supply shock should accelerate a longer investment cycle in energy resilience, as governments shift from consumer protection toward demand reduction, energy security, localized generation, renewables, nuclear power, and stronger grids.
• China reduced crude-oil imports from roughly 11 million barrels per day before the war to an estimated 7.8 million barrels per day in May, while its exports of EVs, lithium-ion batteries, and solar rose 70% year over year in Q1 2026 to a record $21.9 billion.
• Brown Advisory classifies sovereigns by energy position and AI position, using indicators including net oil and gas trade, clean-electricity capacity, grid constraints, data-center interconnections, AI exports, semiconductor exposure, and data-center foreign direct investment.
• The framework places Spain among energy-advantaged AI haves and Italy among energy-disadvantaged AI have-nots; Australia’s data-center investment represented 17% of private investment in Q1, while South Korea’s AI-led growth is concentrated in memory-price gains benefiting Samsung and SK Hynix.

Takeaway: Energy-advantaged countries able to attract AI investment may sustain stronger growth and currencies but structurally higher rates, whereas energy-disadvantaged countries face greater inflation, currency, and stagflation risks; the authors stress that this preliminary framework complements rather than replaces country-level research.