Answers are grounded in this paper and cite their sources.
An environmental conundrum has arisen as a result of the quick development of cloud computing and artificial intelligence. While technology hastens, the world becomes less ecologically sustainable. Data centers that power artificial intelligence use enormous amounts of energy, most of which comes from non-renewable sources. Training advanced artificial intelligence models can even have carbon footprints on par with multiple transatlantic flights. Although some of the largest cloud providers are increasingly buying renewable energy and carbon offsets, those initiatives are nowhere close to keeping pace with our accelerating demands. There are promising new options at our disposal, including carbon-aware computing that schedules workloads to be run when availability is at its lowest, server underclocking, and applying artificial intelligence for load-balancing workloads, which reduces energy usage. It is also an interesting time to integrate FinOps-oriented decision-making with sustainability indicators for responsible cloud governance. These are exciting steps, and they reinforce the fundamentally important transformation we need to see: environmental impact as a key consideration in the design of our digital infrastructures. As the technology sector continues to innovate, it must balance the “social good” associated with AI and cloud functionality alongside the long-term environmental costs and benefits tied to these technologies and their alignment with global sustainable development goals.
No takes yet. Share an insight, caveat, or question.
N. S. Baer (2025) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: