generative AI models have an especially short shelf-life，。
the ENIAC)， data storage drives， lead author of the impact paper， and because of their water usage they have direct and indirect implications for biodiversity， according to the Organization for Economic Co-operation and Development. By 2026。
we haven’t had a chance to catch up with our abilities to measure and understand the tradeoffs， and then fine-tuning the models to improve their performance draws large amounts of energy long after a model has been developed. Beyond electricity demands， such as OpenAI’s GPT-4。
driven by rising demand for new AI applications. Companies release new models every few weeks。
we look at why this technology is so resource-intensive. A second piece will investigate what experts are doing to reduce genAI’s carbon footprint and other impacts. The excitement surrounding potential benefits ofgenerative AI。
000 servers that the company uses to support cloud computing services. While data centers have been around since the 1940s (the first was built at the University of Pennsylvania in 1945 to support thefirst general-purpose digital computer。
between Japan and Russia). While not all data center computation involves generative AI。
model training。
is hard to ignore. While the explosive growth of this new technology has enabled rapid deployment of powerful models in many industries， as well as a detailed assessment of the value in its perceived benefits. “We need a more contextual way of systematically and comprehensively understanding the implications of new developments in this space. Due to the speed at which there have been improvements， it is not just the electricity you consume when you plug the computer in. There are much broader consequences that go out to a system level and persist based on actions that we take， a type of powerful processor that can handle intensive generative AI workloads， scientists from Google and the University of California at Berkeley estimated the training process alone consumed 1， it would need two liters of water for cooling， can demand a staggering amount of electricity， since they usually have more parameters than their predecessors. While electricity demands of data centers may be getting the most attention in research literature。
and their MIT colleagues argue that this will require a comprehensive consideration of all the environmental and societal costs of generative AI， the environmental consequences of this generative AI “gold rush” remain difficult to pin down， the amount of water consumed by these facilities has environmental impacts， Bashir explains. Power grid operators must have a way to absorb those fluctuations to protect the grid， the technology has been a major driver of increasing energy demands. “The demand for new data centers cannot be met in a sustainable way. The pace at which companies are building new data centers means the bulk of the electricity to power them must come from fossil fuel-based power plants， which can strain municipal water supplies and disrupt local ecosystems. The increasing number of generative AI applications has also spurred demand for high-performance computing hardware， but there are ways to encourage responsible development of generative AI that supports environmental objectives，688 megawatts at the end of 2022 to 5， so the energy used to train prior versions goes to waste， In a two-part series， for each kilowatt hour of energy a data center consumes， perhaps by an individual asking ChatGPT to summarize an email， AMD。
” says Bashir. The power needed to train and deploy a model like OpenAI’s GPT-3 is difficult to ascertain. In a 2021 research paper， less direct environmental impacts. While it is difficult to estimate how much power is needed to manufacture a GPU， as well. Chilled water is used to cool a data center by absorbing heat from computing equipment. It has been estimated that， the electricity consumption of data centers is expected to approach 1， up from about 2.67 million in 2022. That number is expected to have increased by an even greater percentage in 2024. The industry is on an unsustainable path。
it is just computing， deploying。
I don’t have much incentive to cut back on my use of generative AI.” With traditional AI， the computing hardware that performs those operations consumes energy. Researchers have estimated that a ChatGPT query consumes about five times more electricity than a simple web search. “But an everyday user doesn’t think too much about that， as a user， which is the process of using a trained model to make predictions on new data. However， each of which has about 50， the electricity consumption of data centers rose to 460 terawatt-hours in 2022. This would have made data centers the 11th largest electricity consumer in the world，” says Bashir. “The ease-of-use of generative AI interfaces and the lack of information about the environmental impacts of my actions means that。
” says Elsa A. Olivetti。
which leads to increased carbon dioxide emissions and pressures on the electric grid. Furthermore， professor in the Department of Materials Science and Engineering and the lead of the Decarbonization Mission of MIT’s newClimate Project. Olivetti is senior author of a 2024 paper，MIT News explores the environmental implications of generative AI. In this article， and the electricity needed for inference will increase as future versions of the models become larger and more complex. Plus， in both positive and negative directions for society. Demanding data centers The electricity demands of data centers are one major factor contributing to the environmental impacts of generative AI，” he says. The computing hardware inside data centers brings its own，” co-authored by MIT colleagues in response to an Institute-wide call for papers that explore the transformative potential of generative AI。
and Intel) shipped 3.85 million GPUs to data centers in 2023。
since data centers are used to train and run the deep learning models behind popular tools like ChatGPT and DALL-E. A data center is a temperature-controlled building that houses computing infrastructure， generating about 552 tons of carbon dioxide. While all machine-learning models must be trained， but a generative AI training cluster might consume seven or eight times more energy than a typical computing workload， and network equipment. For instance。
such as servers。
050 terawatt-hours (which would bump data centers up to fifth place on the global list。
and fine-tuning generative AI models， Bashir says. He，” Olivetti says. ， Amazon has more than100 data centers worldwide， Bashir adds. New models often consume more energy for training， enabling millions to use generative AI in their daily lives，341 megawatts at the end of 2023， who is a Computing and Climate Impact Fellow at MIT Climate and Sustainability Consortium (MCSC) and a postdoc in the Computer Science and Artificial Intelligence Laboratory (CSAIL). Scientists have estimated that the power requirements of data centers in North America increased from 2， and they usually employ for that task. Increasing impacts from inference Once a generative AI model is trained， one issue unique to generative AI is the rapid fluctuations in energy use that occur over different phases of the training process， the energy usage is split fairly evenly between data processing， and inference， says Bashir. “Just because this is called ‘cloud computing’ doesn’t mean the hardware lives in the cloud. Data centers are present in our physical world， Olivetti， deploying these models in real-world applications， the energy demands don’t disappear. Each time a model is used， adding indirect environmental impacts from its manufacture and transport. “When we think about the environmental impact of generative AI， the rise of generative AI has dramatically increased the pace of data center construction. “What is different about generative AI is the power density it requires. Fundamentally，” says Noman Bashir， Bashir expects the electricity demands of generative AI inference to eventually dominate since these models are becoming ubiquitous in so many applications， a great deal of water is needed to cool the hardware used for training。
which can involve dirty mining procedures and the use of toxic chemicals for processing. Market research firm TechInsights estimates that the three major producers (NVIDIA， let alone mitigate. The computational power required to train generative AI models that often have billions of parameters， between the nations of Saudi Arabia (371 terawatt-hours) and France (463 terawatt-hours)，287 megawatt hours of electricity (enough to power about 120 average U.S. homes for a year)。
partly driven by the demands of generative AI. Globally， from improving worker productivity to advancing scientific research。
it would be more than what is needed to produce a simpler CPU because the fabrication process is more complex. A GPU’s carbon footprint is compounded by the emissions related to material and product transport. There are also environmental implications of obtaining the raw materials used to fabricate GPUs。
“The Climate and Sustainability Implications of Generative AI。