Generative artificial intelligence (GenAI) has rapidly transformed how we interact with machines, particularly through natural language processing (NLP). From chatbots and automated writing assistants to creative tools and research companions, GenAI is increasingly embedded in our daily digital lives. These tools, like OpenAI\u2019s ChatGPT<\/a>, Google\u2019s Gemini<\/a>, Anthropic\u2019s Claude<\/a>, and DeepSeek<\/a>, are powered by large language models (LLMs) that generate human-like responses based on prompts we give them. <\/p>\n\n\n\n How these systems understand us depends on their ability to process natural language, breaking down our words into manageable, computationally understandable pieces called tokens. The ease with which we can now generate responses using GenAI contradicts the immense computational and environmental power behind each interaction.<\/p>\n\n\n\n The way AI models process our prompts has an impact on the environment. When you type a question into a LLM like ChatGPT, it does not read your message word-for-word in the way humans do. Instead, it breaks down your input into tokens: small fragments of text, typically chunks of 3 – 4 characters or individual words. For instance, the sentence \u201cThink before you prompt\u201d may be split into six tokens. Each token processed consumes computational resources, requiring both energy and water for the massive data centres that power these models. <\/p>\n\n\n\n It\u2019s not just a technical detail, it\u2019s a sustainability concern.<\/em><\/p>\n\n\n\n According to the World Economic Forum (2024)<\/a>, AI can help us tackle climate change: it can predict the weather, track icebergs and identify pollution. It can also improve agricultural output and reduce its environmental impact. However, engaging with LLMs carries an environmental footprint that needs ethical regulation and greener tech to prevent risks and inequalities (UNU EHS, 2024<\/a>). <\/p>\n\n\n\n Let\u2019s break down the environmental cost of an interaction with a GenAI LLM:<\/p>\n\n\n\n GenAI also increases the need for resource-heavy hardware like Graphics Processing Units (GPUs) or Tensor Processing Units (TPUs). Producing these components requires rare-earth metals, whose extraction often causes environmental harm and high emissions, adding to AI\u2019s overall ecological impact (Walther, 2024<\/a>).<\/p>\n\n\n\n It\u2019s not just about AI efficiency anymore\u2014it’s about digital sustainability. And the more iterative or unclear your prompts, the more resources are used<\/em> (CoDesignS AI Framework, 2024<\/em><\/a>).<\/p>\n\n\n\n Let\u2019s have a closer look at the environmental impact of some of the leading LLMs.<\/p>\n\n\n\n As presented in Table 1 below, environmental impact is strongly influenced by the LLM\u2019s token input capacities and efficiency. Models with large token limits, such as Gemini 1.5 Pro and GPT-4o, require significant computational power, leading to higher energy consumption. In contrast, DeepSeek V3 claims to be more environmentally sustainable, using only about one-tenth of the computational resources compared to similar models, although this is yet to be independently verified (Calma, 2025<\/a>). Claude 3.7\u2019s environmental performance remains unclear due to limited transparency. Overall, while handling larger contexts improves capabilities, it also increases environmental costs unless offset by greater model efficiency.<\/p>\n\n\n\n Every additional prompt adds more tokens to the pile\u2014more energy, more water, more emissions. It\u2019s tempting to view interactions with LLMs as \u201cfree,\u201d but behind the scenes, there is a high cost.<\/em><\/p>\n\n\n\n This brings us to a pivotal mindset shift: less is more. Instead of multiple iterations and retries, the goal should be to get the best response in fewer interactions. Every refined, thoughtful prompt saves resources and time.<\/p>\n\n\n\n Prompt engineering is the skill of crafting clear, effective inputs to get the desired outputs from GenAI models. It\u2019s like learning to ask the right question in the right way. The better the prompt, the better (and faster) the answer.<\/p>\n\n\n\n Prompt engineering is an essential tool in reducing GenAI\u2019s environmental footprint (CoDesignS AI Framework, 2024<\/a>). It reduces trial and error, minimises redundant queries, and shortens conversation length. Not only does this save money and computing time, but it directly reduces electricity and water usage.<\/p>\n\n\n\n But how do we create better prompts? This is where a framework like ROCKS comes into play.<\/p>\n\n\n\n Developed as part of the open access CoDesignS AI Framework (2024)<\/a>, ROCKS is a practical, easy-to-remember method to help us optimise our interactions with GenAI systems. <\/p>\n\n\n\n As presented in Figure 2, ROCKS stands for:<\/p>\n\n\n\n Role<\/strong>: Identify your role. For example, instead of writing:<\/p>\n\n\n\n \u201cCan you help me make my lecture more interactive?\u201d<\/em><\/p>\n\n\n\n Use the ROCKS structure for greater clarity and focus, leading to fewer follow-up questions and lower energy consumption:<\/p>\n\n\n\n Role<\/strong>: I am a lecturer in Pathology teaching Year 2 MBBS students. <\/p>\n\n\n\n Objective<\/strong>: I aim to make my lecture on the Pathology of the Head more interactive, dynamic, and enjoyable to better engage students during the session. <\/p>\n\n\n\n Community<\/strong>: The lecture will be delivered in-person to around 300 Year 2 MBBS students in a large lecture theatre setting. <\/p>\n\n\n\n Key\/Tone<\/strong>: The tone should be engaging, energetic, and inclusive, encouraging active participation and maintaining attention across a large group. <\/p>\n\n\n\n Shape<\/strong>: I would like suggestions for practical, easy-to-implement methods using PowerPoint and Kahoot, focusing on live interaction, quizzes, and gamified learning elements that are scalable for a large cohort.<\/p>\n\n\n\n Using the ROCKS method, you can craft more precise and purposeful prompts that not only improve the quality of AI responses but also contribute to more efficient and sustainable GenAI use.<\/p>\n\n\n\n GenAI is transforming how we think and work, but every interaction draws on the planet\u2019s finite resources. By learning to prompt carefully and with purpose, we can reduce wasteful iterations, improve response quality and minimise the environmental impact of our digital habits. Tools like the ROCKS method empower us to be not just efficient, but ethical. As educators, learners and professionals, we have a responsibility to pair innovation with sustainability. The next time you engage with an AI, pause to refine your prompt; not just for a better answer, but for a better future.<\/p>\n\n\n\n Think before you prompt\u2014because better inputs create better outputs for us and a better future for the planet (<\/em>CoDesignS ESD AI Coach, 2024<\/em><\/a>).<\/em><\/p>\n\n\n\n Calma, J. (2025) \u2018AI is \u2018an energy hog,\u2019 but DeepSeek could change that\u2019. The Verge. January. Available at:<\/p>\n\n\n\n https:\/\/www.theverge.com\/climate-change\/603622\/deepseek-ai-environment-energy-climate?utm_source=chatgpt.com<\/a> last accessed: 7 March 2025<\/p>\n\n\n\n CoDesignS ESD AI Coach (2025). Available at: https:\/\/codesignsesd.org\/codesigns-ai-coach\/<\/a> last accessed: 7 March 2025<\/p>\n\n\n\n CoDesignS AI Framework (2024). Available at: https:\/\/aldesd.org\/7574-2\/<\/a> last accessed: 7 March 2025<\/p>\n\n\n\n Cho, R. (2023). AI\u2019s Growing Carbon Footprint. June. Columbia Climate School Newsletter. Available at:<\/p>\n\n\n\n https:\/\/news.climate.columbia.edu\/2023\/06\/09\/ais-growing-carbon-footprint\/<\/a> last accessed: 7 March 2025<\/p>\n\n\n\n GPAI \u2013 The Global Partnership on Artificial Intelligence (2021). \u2018Climate Change and AI \u2013 Recommendations for Government action\u2019. Available at: https:\/\/www.gpai.ai\/projects\/climate-change-and-ai.pdf<\/a> last accessed: 7 March 2025<\/p>\n\n\n\n Li, P., Yang, I., M., Ren, S. (2023) \u2018Making AI Less “Thirsty”: Uncovering and Addressing the Secret Water Footprint of AI Models\u2019 https:\/\/doi.org\/10.48550\/arXiv.2304.03271<\/a> <\/p>\n\n\n\n Luccioni, S., Jernite, Y., Strubell, E. (2024). Power Hungry Processing: Watts Driving the Cost of AI Deployment? In Proceedings of the 2024 ACM Conference on Fairness, Accountability, and Transparency (FAccT ’24). Association for Computing Machinery, New York, NY, USA, 85\u201399. https:\/\/doi.org\/10.1145\/3630106.3658542<\/a><\/p>\n\n\n\n Masterson, V. (2024). \u20189 ways AI is helping tackle climate change\u2019. Blog post. February. Available at: https:\/\/www.weforum.org\/stories\/2024\/02\/ai-combat-climate-change\/<\/a> last accessed: 7 March 2025<\/p>\n\n\n\n UNU EHS (2024). \u20185 Insights into AI as a Double-Edged Sword in Climate Action\u2019. Blog post. June. Available at: https:\/\/unu.edu\/ehs\/series\/5-insights-ai-double-edged-sword-climate-action<\/a> last accessed: 7 March 2025<\/p>\n\n\n\n US Environmental Protection Agency. 2024. Greenhouse Gases Equivalencies Calculator – Calculations and References. https:\/\/www.epa.gov\/energy\/greenhouse-gases-equivalencies-calculator-calculations-and-references<\/a> last accessed: 7 March 2025<\/p>\n\n\n\nHow do AI models process our prompts?<\/h2>\n\n\n\n
The environmental footprint of a prompt<\/h2>\n\n\n\n
\n
Comparing Leading LLMs: Environmental impact<\/h2>\n\n\n\n
![\"A](\"https:\/\/i0.wp.com\/altc.alt.ac.uk\/blog\/wp-content\/uploads\/sites\/1112\/2025\/05\/AD_4nXd_CJhOjXEl8CWgAZTQ0Gq5HxuFXDbIa_U4C5TN75QzY5EmvLLF_DUMTGiyef5pKNfLV-sKFtYaj3-ByB5mCXz075oImlV1O4pSnL9KMfp2uW1TfWIAFIBgIK-Ozy3TqQOgrMHeLA.png?resize=1378%2C747&ssl=1\")
Less is more: Why every prompt matters<\/h2>\n\n\n\n
Optimise your interactions: Introducing the ROCKS method<\/h2>\n\n\n\n
![\"A](\"https:\/\/i0.wp.com\/altc.alt.ac.uk\/blog\/wp-content\/uploads\/sites\/1112\/2025\/05\/AD_4nXcQ1drWTBUehAlsbQdMvZB_yqq0kQQb5GDmV3aRZw7TnVj_ofet8Pxqkp0McZgJ0oCjbAeKeATfOKx_GMcQCs77I5uiEMsADiY2uLkADtaq7gtwc67rUTt8nrk17HOZ5gtkBxi7jQ.png?resize=799%2C797&ssl=1\")
Objective<\/strong>: State your objective.
Community<\/strong>: Specify your audience
Key<\/strong>: Describe the tone or style, and any related parameters.
Shape<\/strong>: Note the desired format of the output.<\/p>\n\n\n\nAn example of a prompt using the ROCKS method<\/h3>\n\n\n\n
Conclusion: Prompt with Purpose<\/h2>\n\n\n\n
REFERENCES<\/h2>\n\n\n\n