AI Expansion Could Cause 2.8 Gigatons of CO₂ by 2031

A recent study predicts that the expansion of artificial intelligence (AI) could lead to 2.8 gigatons of CO₂ emissions by 2031 if these technologies are not integrated into more environmentally friendly processes. The research highlights that the energy consumption of AI applications significantly contributes to the global CO₂ balance. The study analyzes various aspects of AI expansion, including the data centers required for training AI models. These data centers consume large amounts of energy, leading to an increase in greenhouse gas emissions. The focus is on the necessity of utilizing sustainable energy sources to minimize environmental impacts.

A key finding of the investigation is that the current trend of using AI across various sectors, such as healthcare, automotive, and financial services, is unsustainable without appropriate measures to reduce energy consumption and emissions. The researchers therefore call for a comprehensive strategy to integrate renewable energy into the AI sector. The study points out that implementing environmentally friendly technologies and processes is crucial to mitigate the negative impacts of AI expansion on the climate. This includes, among other things, using solar and wind energy to power data centers and developing energy-efficient algorithms. Another aspect of the study is the role of governments and companies.

They are urged to create clear guidelines and incentives to promote the transition to more sustainable practices in the field of artificial intelligence. The researchers emphasize that without such measures, the climate goals of most countries are at risk. The discussion about the environmental impacts of AI is not new, but it is gaining urgency as the technology becomes increasingly integrated into daily life. Experts warn that time is running out to find suitable solutions before emissions rise uncontrollably. The study was conducted by an international team of scientists and published in a renowned scientific journal.

The findings are intended to serve as a basis for future policy decisions and corporate strategies to reduce the ecological footprint of AI technologies. A concrete example of AI's energy consumption is the training of large language models, which requires several megawatt-hours of electricity. These figures underscore the need to monitor and optimize energy consumption in AI development. The researchers appeal to the industry to pursue innovative approaches to enhance the efficiency of AI systems.

This could be achieved through the use of edge computing, where data is processed closer to its source to reduce energy demand. The study concludes by stating that the expansion of artificial intelligence brings both opportunities and challenges. The coming years will be crucial in setting the course for sustainable development in this area. The researchers emphasize that measures must be taken by 2031 to avoid the projected 2.8 gigatons of CO₂.