Microsoft’s ongoing efforts to address the environmental impact of AI and cloud computing through investments in sustainability and renewable energy, despite challenges such as increased emissions from data infrastructure expansion.
Microsoft’s recent initiatives highlight the environmental impact of AI and the efforts to mitigate it. In 2020, Microsoft pledged to be carbon negative by 2030 and to offset all historical emissions by 2050. However, the company’s emissions increased by 29% since 2020 due to massive investments in data infrastructure.
This week, Microsoft announced a $3.2 billion investment to expand its cloud computing in Sweden, part of a broader $50 billion expenditure on data centres this year. They also plan up to a $100 billion investment in US-based supercomputers and data centres with OpenAI.
Data centres hosting AI models consume significant energy and water. According to the International Energy Agency, these centres accounted for less than 1.5% of global electricity use in 2022 but could reach 1,000 terawatt hours by 2026, equivalent to Japan’s electricity use.
Microsoft aims to improve sustainability by enhancing AI energy efficiency and investing in renewable energy. Last month, they committed $10 billion to bring 10.5 gigawatts of renewable energy online in partnership with Brookfield Asset Management. Additionally, they are exploring small nuclear modular reactors and have agreed to purchase nuclear fusion-generated electricity from Helion Energy by 2028.
Researchers and companies also explore AI applications in weather prediction and geoengineering to address climate change impacts. In November, Google DeepMind introduced GraphCast, an AI model for more accurate and energy-efficient 10-day weather forecasts.
While AI’s environmental costs are clear, its potential benefits in tackling climate challenges remain to be fully realized.
