How does generation AI have an impact on the environment?
Artificial intelligence has generated quite a lot of talk over the past few years. We talked about its outlook and immeasurable benefits, and even potential consequences of ethical use, moving jobs, and eliminating certain transactions. But what aspects do most of us consider at the current stage of development? What do you know about the environmental impact of AI?
Before looking for AI tools to generate professional headshots for LinkedIn, let’s review our current research. Is the footprint worth it? What can we do to use this exciting and fresh technique as carefully as possible?
There is everything you need to know about generative AI and its environmental impact. This article is not intended to prevent the use of AI. Rather, it is incentive to launch an open conversation about how we can benefit from it and ensure that safety measures are in place for our planet.
Environmental impact of AI generated: Numbers and statistics
It’s safe to say that while many people were excited when the first AI model became publicly available, others didn’t. Witnessing a technical breakthrough in your life has been an unparalleled experience, which has led many people to try out chatbots and genai tools. But what preceded their launch and what does that mean for the environment? Let’s talk about the numbers.
AI techniques and tools start with training. Simply put, this means supplying the data needed to provide answers to the AI model. Training a single AI model consumes thousands of hours of megawatts. GPT-3 training requires 1,287 MWh and emits 552 tons of CO2. [1] This is equivalent to 123 gasoline-powered passenger cars that are powered for a year. [2] Additionally, electricity fluctuates during the training phase, and a stability solution for the power grid, which often uses diesel-based generators, is required. By combining it with energy usage, it is safe to say that the environmental footprint is large through deployment beyond training, user interaction and repetition.
Furthermore, data centers are central to training and execution of AI models. Typically, each contains tens of thousands of servers, and they are growing rapidly to keep up with the growing demand for AI. From 2022 to 2023, the energy needs of data centers in North America doubled from 2,688 MW to 5,341 MW. In 2022, global use reached 460 TWH. By 2026, global data center consumption could reach TWH of 1,050, marking fifth among the world’s largest electricity consumers. [1]
Additionally, data centers require cooling towers and air mechanisms to dissipate heat. This requires a significant amount of water. A rare and finite natural resource, AI’s water footprint also affects local water supply, ecosystems and biodiversity. [1]
As AI technology advances further, it is best to think that the environmental impact will decrease, become more manageable, and ultimately the design will become more sustainable. For now, however, new models are usually large, which exacerbate the cycle of increased energy demand, and there is still a significant data gap to measure actual short- and long-term environmental impacts. The expansion of AI’s daily life and integration into apps can lead to unconscious overuse by everyday users, making it essential to establishing more mindful and sustainable usage parameters.
How can you use the generated AI more mindfully?
The more sustainable use of AI development, deployment and use results in a more sustainable fall primarily for those who create it. However, as an end user, when you are going to use these tools as effectively and heartfelt as possible in your professional and personal life, you can set guidelines for yourself.
Educate yourself
Now, you need to use AI tools to keep your administrative tasks out of the way, or streamline your processes even further. This is a good opportunity to learn more about AI in general. As with most things, educating yourself on this subject does better than harm. Take more time to learn what is needed to develop and train your AI models. Find out as much as possible about the data you use and the methods that are programmed to process it. Learning more about this topic is that staying in the spotlight to the next big thing is an investment in becoming a more responsible user.
Is there another way to get your answer?
Can you ask a colleague, friend, or expert about what you need to know? Can I search through other means, such as web search or online communities? Yes, internet searches have their own environmental impact. At the end of 2023, Google’s greenhouse gas emissions (GHG) amounted to 14.31 million tonnes of carbon dioxide. [3] However, a significant proportion of this figure was primarily due to the expansion of data centers needed to operate and expand AI services.
Research also estimates that a single prompt to ChatGPT would require five times more power than a web search. [1] However, as many search engines now embed AI answers in their search results, the future shift in how the internet is used is certainly interesting, not to mention new numbers on environmental footprints. Still, before sending a quick question to your favorite chatbot, consider whether it’s worth investing extra effort to find answers elsewhere.
Promote knowledge sharing
Read on recently about high-quality works on the environmental impact of AI and the unprecedented leap in its technology. Share it with your friends. If your company requires that AI tools be integrated into your everyday workflow, set up a training session that addresses your concerns and provides insight into the inference behind your implementation. Talk to an expert with AI expertise to educate your employees on the complexity of this technology and how to fully utilize it. Or, track these new trends and advances and bring together teams that will give the rest of your organization a scoop. By establishing an effective knowledge sharing channel, you (and your team or company) can learn how to make the most of these new technologies and decide how you can get the returns you want.
(Try it) Balance
Because many of us use AI. Because it’s to make things easier, or simply because we want to explore this technical breakthrough happening in front of us. All of these are reasons why they are completely accepted. Still, if you know you’re going to use a Genai tool that increases frequency, try returning something to your environment. The numbers and environmental impacts behind AI use were seen. Mere individuals can’t overcome the environmental impact of the huge data centers they work to keep AI up and running, but they can try to be environmentally sensitive. Practice sustainability, establish an organization-wide program for corporate environmental responsibility, work from home, get bikes, plant some trees, balance just a little bit of scale. It may be an idealistic effort, but collective effort can ultimately make a difference.
Conclusion
At this point, almost everyone is using AI with some abilities and it’s great to try something new. Its innovative possibilities are certainly worth exploring. The key is to form your own opinion on when and how to proceed. Do you want AI to be a tutor, assistant, search engine, recipe developer, or something else? it’s up to you! Generic AI has the power to disrupt the world, as we know it. Its impact reaches beyond the redefine efficiency and covers almost every aspect of human activity. It all starts in our environment. It is more important than ever that we are seeking educational opportunities to remain mindful and to deepen our understanding, and of course, we are playing our part in ensuring a better tomorrow for future generations.
References:
[1] Description: The impact of generation AI on the environment
[2] Computer scientists break down the massive carbon footprint of generated AI
[3] Net Zero Carbon
