Blog
09.05.2023

Last month, people around the world gathered to celebrate Earth Day, the annual event to demonstrate support for environmental protection. This year’s theme, “Invest in Our Planet”, focused on engaging governments, businesses and individuals to make environmentally-conscious decisions that prioritise the health of our planet and future generations. Investing in our planet means investing in new technologies that optimise our sustainability efforts and ensure efficiency and accuracy in everything we do. New avenues for applications of Artificial Intelligence (AI) and high performance computing (HPC) in green projects continue to open up as the technology grows more intuitive and creative. 

The intelligence of AI has developed to the extent that its artificial component is becoming undetectable. This refers specifically to generative AI – AI that processes large quantities of data to create original content, including code, text, images, videos and simulations. Generative AI was popularised last year by the launch of OpenAI’s chatbot ChatGPT, which set the record for the product with the fastest growing user-base. Generative AI has proven the creative capabilities of AI, something we previously underestimated, advancing its role in sustainability efforts from background analytics to being the face of campaigns. In advance of Earth Day 2023, WWF released an exhibition of AI-generated images envisioning a bleak future for UK nature without environmental protection. The exhibition, entitled Future of Nature, features apocalyptic scenes in the style of British Romantic artwork, deliberately making irony of the Romantic focus on the beauty of nature. The artwork sends the powerful message that the fate of our world is in our hands, linking to the futuristic theme of Earth Day 2023. 

Generative AI also has a role to play in designing the cities of the future. The architecture company Maket utilises AI to enhance building design and urban planning. Maket’s technology can generate 3D renderings of buildings and cities that prioritise energy efficiency, protection of green space and local ecology and reduction of waste and pollution for better quality of life in urban environments. Experts forecast that the future of generative AI will be in specialised domains rather than general-purpose use. Niche applications, such as realizing an artistic concept or designing buildings, can benefit from the ability of generative AI to generate unique and optimised solutions, whereas generalised applications may be less effective in the face of multiple tasks and large quantities of data.

Edge computing is a rapidly growing field that lends itself well to the harnessing of data for hyperlocal climate research. Researchers from Northwestern University and Argonne National Laboratory have been launching edge AI-driven Waggle sensors around the world to create custom climate models that can provide more precise and localised predictions of weather patterns and climate change. Edge computing based environmental monitoring has various useful applications, such as studying wildfire patterns in forests and observing heat waves in urban areas. It is particularly useful for wildlife conservation efforts because, thanks to reduced latency and bandwidth requirements, conservationists can respond to threats to endangered species, such as poachers, in real-time and therefore protect wildlife more efficiently. Conservation AI has deployed 70+ AI-powered cameras worldwide, trained via deep learning to detect and report such threats. By adapting our digital infrastructure, we can preserve our planet’s most precious ecosystems.

The transition to a net zero economy will require significant investments in renewable energy, but its drawbacks, namely variability and unpredictability of energy output, prolong our reliance on fossil fuels. AI is helping us manage these areas of uncertainty. Vestas, a leading wind turbine manufacturer, is utilising generative AI to maximise the power output of wind turbines by adjusting blade angles and other factors in real-time, based on wind speed and direction, a method known as wake steering. The technology can also select optimal locations for wind farms and predict when turbine maintenance is needed, enabling more effective and sustainable energy generation to progress the renewable revolution.

As the intelligence and capability of AI grows, the demand for computing power to support new applications increases dramatically. These powerful tools that support our sustainability efforts are also major contributors to the problem without the backing of renewable energy at every step of their development. It is more important than ever to scrutinise the carbon footprint of upcoming technology to ensure that we are not just bringing more shiny new toys into the world – rather, we are prioritising environmental impact and making meaningful investments in our future and the future of our planet.