AI surge strains energy-hungry data centres
Smarter infrastructure and policy support urged as data storage demands soar.
As artificial intelligence accelerates global data generation, pressure is mounting on data centres to meet both performance and sustainability expectations. In Singapore, 90% of professionals are already concerned about the environmental cost of data storage, according to Seagate’s Decarbonising Data report.
Joyce Lim, ASEAN Business Lead at Seagate Technology, outlined a three-pillar strategy to reduce the environmental footprint: embrace technological innovation, extend product life cycles through circular practices, and ensure ecosystem-wide collaboration.
“Technological innovation will be the key driver for sustainable transformation,” she said, citing Seagate’s Mozaic 3+ platform that “increased capacity by three times,” reduced embodied carbon by over 70%, and cut costs by 25%.
She added that adopting tiered storage infrastructure—using high-performance SSDs for frequently accessed data and power-efficient drives for archival storage—can significantly lower power use. Joyce also highlighted that data centres can also address space constraints by transiting from 10TB hard drives to 30TB hard drives. “You are able to increase capacity by three times... and reduce power consumption by 60% without taking up more space.”
Dr Sean Shin, Senior Lecturer at NUS Business School, warned that AI is intensifying energy demands.
“A single ChatGPT search can be up to 10 times more energy intensive than a typical Google search,” Shin said, noting a projected 33% annual increase in demand for AI-ready data center capacity until 2030.
While renewable energy is often promoted as a solution, Shin said implementation is complex in space-constrained Singapore. He pointed to innovations like AI-driven and liquid cooling systems, as well as carbon offsets.
Governments, both experts agreed, have a critical role.
“The government can incentivise green tech through grants, tax breaks, and R&D funding,” Lim said, referencing Singapore’s sustainable tropical data center testbed.
Shin added, “Policies like the green data center roadmap ensure smart, green growth... they are the guardrails that can accelerate real progress.”
Commentary
Singapore’s AI ambition has an intelligence problem
Singapore is deploying AI at speed – and security risks are catching up
Precious metal boom is a stress test for Singapore’s gold ambitions
Singapore’s global dispute fault lines
‘Tokenmaxxing’ – The wrong AI race to run in Singapore
To outsmart modern fraud, we must first know the enemy
Why Singapore SMEs cannot wait for quantum cyber risk to arrive before securing data