Top tips on using generative AI tools in an environmentally friendly way

Since the use of generative AI tools like Copilot and ChatGPT has surged, many concerns have been raised around their impact on the environment. While offering huge productivity and creativity benefits, tools like these rely on vast data centres that consume significant amounts of energy. Currently, these data centres consume between 2% and 3% of the world's electricity, according to the International Energy Agency. With Goldman Sachs Research forecasting demand from data centres is set to increase 50% by 2027 and by as much as 165% by the end of the 2020s.  

As more people and organisations integrate AI into their ways of working, the environmental footprint of these technologies becomes increasingly difficult to ignore. Big tech companies are facing mounting pressure too. With the media, investors, customers, and the general public demanding positive action to offset the impact of AI on the environment. 

In this article, we’ll explore how individuals can help. Providing six top tips for using generative AI tools in a more environmentally responsible and sustainable way. 

 

 

1. Write clear and concise prompts 

By writing clear, concise and direct prompts when using generative AI tools like ChatGPT and Copilot, users reduce their environmental impact by minimising the effort required to generate accurate responses.  

When prompts are vague or overly complex, the AI tools may need to process more data or generate longer outputs, which increases energy consumption. Clear prompts lead to faster, more efficient interactions, reducing the number of follow-up queries and the overall carbon footprint of each session. 

You can read our guide to writing good AI prompts here, for more tips and tricks on this!  

 

 

2. Batch your questions and queries

Batching your questions and queries within a single session also contributes to lower environmental impact, by reducing the frequency of server actions and data processing. Instead of repeatedly asking the AI tool to complete individual tasks, grouping related questions allows the systems to operate more efficiently, helping to conserve the energy required. 

 

 

3. Use search engines for some queries 

It’s important to use the right tool for the right task. For straightforward queries like checking the weather, finding a definition, or looking up a fact, using a traditional search engine can be significantly more energy-efficient than generative AI tools like ChatGPT or Copilot. Which are, generally, far better suited for more complex queries or requests.  

Generative AI models require far more power to process and generate responses, with a request made through ChatGPT consuming 10 TIMES the electricity of a Google Search, according to the International Energy Agency. 

 

 

4. Limit image generation 

We’ve all seen the fun AI image trends on social media, like the action figure box creations that went viral around April 2025. While it can be good fun to jump on trends like this, (we did ourselves! Check it out on LinkedIn) it’s important to keep in mind that generating an image like this requires huge energy resource; much more than a simple generative AI text response request.  

According to a new study by researchers at Hugging Face and Carnegie Mellon University, generating images through AI tools takes as much energy as fully charging your smartphone. While using these tools to generate text is significantly less energy-intensive; equal to charging your phone to around 16%. 

To use AI in an environmentally responsible way, it’s important to limit the number of images generated. Each time you do it, it’s having a significant environmental impact!  

 

5. Close the application when you’re done 

When you’ve used your generative AI tool and got the answer or result you’re looking for, don’t forget to close down the application. Doing so can help preserve energy, by reducing server load and minimising idle processing. While the energy savings from a single person doing this may not be huge, collective action across millions of users around the world can make a huge, meaningful impact. 

So – close that ChatGPT or Copilot tab you’ve got open all day!  

 

 

6. Use AI with purpose 

Ultimately, the best way to use generative AI in an environmentally responsible way is to use it with purpose.  

Ensure you are avoiding unnecessary or repetitive queries that consume energy without adding value. Be intentional about when and how you engage with AI tools. And, remember that each interaction you have with AI contributes to a much larger energy footprint.  

Purposeful use not only reduces environmental impact but also encourages more thoughtful, effective engagement with AI overall. 

 

 

Generative AI use and the environment – Final thoughts  

While generative AI offers – and indeed is already delivering – huge productivity and efficiency benefits to individuals and organisations, we should not ignore the serious environmental implications these tools have. As adoption of generative AI accelerates, so too does the carbon it emits.  

Tech companies must lead by example, investing in cleaner infrastructure, transparent reporting, and greener AI practices. However at the same time, individuals can play a role by using these tools thoughtfully and responsibly. 

With the right behaviours, we can all ensure that the transformational benefits of generative AI are not outweighed by its environmental costs.