Artificial Intelligence and Its Impact on the Environment: Hidden Threats of the Digital Revolution

Artificial intelligence (AI) is playing an increasingly significant role in our lives every year. From algorithms managing social networks to autonomous vehicles, AI promises to make the world more efficient, convenient, and connected. However, behind these technological breakthroughs lies another, less obvious side: a significant negative impact on the environment.

One of the main problems associated with AI is its enormous energy consumption. Training complex machine learning models requires powerful computational resources that consume massive amounts of electricity. Training a single large neural network can require hundreds or even thousands of graphics processors working around the clock for several days or weeks. This leads to large carbon dioxide (CO2) emissions, especially if the energy comes from non-renewable sources such as coal power plants. According to a study conducted by scientists from the Massachusetts Institute of Technology (MIT), training a single large machine learning model can produce up to 284 tons of CO2, which is equivalent to five car trips around the Earth. According to the same study, at Google, 56% of energy comes from renewable sources, at Microsoft — 32%, at Amazon AWS — 17%, with the remaining energy coming from coal, oil, and nuclear power. This level of emissions raises serious questions about the environmental sustainability of the technologies shaping our future.

The development of AI and machine learning is also associated with an increase in electronic waste. The equipment used for training AI becomes obsolete faster than many other types of technology. Servers, graphics processors, and other components of AI systems need to be replaced every few years to maintain high performance and remain competitive. Obsolete devices often end up in landfills, where they not only take up space but also pollute the environment with toxic substances like mercury, lead, and cadmium.

One of the less discussed aspects of AI’s negative environmental impact is the use of water. Massive data centers require vast amounts of water to cool their servers. The Washington Post, in collaboration with scientists from the University of California (UC), found out how much water and energy ChatGPT from OpenAI uses to cool its running servers. For a single one-hundred-word email, half a liter of water and 0.14 kilowatt-hours (kWh) of electricity are required, which is equivalent to running fourteen LED bulbs for an hour.

Despite the serious issues related to AI’s environmental footprint, there are ways to mitigate its impact on nature. First, the share of renewable energy in the operation of data centers and computational resources needs to be increased. Many companies have already started transitioning to “green” energy, and this direction should become a priority for all industry players. Second, developing more efficient algorithms can significantly reduce energy consumption. Modern approaches to AI development should focus not only on improving accuracy and performance but also on reducing the resources required for their operation. Finally, recycling and reusing equipment can also reduce the amount of electronic waste. Linara Khadimullina, a specialist in scientific-methodological support at AIM Carbon, emphasizes that there is currently no simple solution to this problem, but it is essential to encourage responsible consumption and raise public awareness of this issue.

“Artificial intelligence can have both positive and negative impacts on the environment. On the one hand, it helps optimize production processes, leading to reduced emissions into the atmosphere. On the other hand, the widespread use of artificial intelligence requires significant energy consumption. It’s important to strike a balance: if companies transparently improve their processes and technologies, then the use of artificial intelligence will be mostly beneficial,” she said.

Artificial intelligence promises to revolutionize many industries and change our world for the better. However, behind this progress lies a serious environmental cost. For AI to truly benefit humanity, it is necessary to consider not only its potential capabilities but also its impact on nature. It is crucial to develop AI technologies while remembering the need to preserve the planet for future generations.