COP28: Accelerating Net Zero Goals with Generative AI
Generative AI will be a key in the fight to save our planet.
Another year another UN Climate Change Conference. This year’s COP28 — scheduled from 30 November to 12 December — is located in Dubai, and aims to drive “global transformation towards a low-emission and climate-resilient world” and foster an “ambitious climate action” that “facilitates implementation, including the related support”.
So far, so COP. However, no matter how cynical you may feel towards the event itself — Greta Thunberg herself opted out of COP last year — the fact of the matter is that the event is very good at showing what can be done as much as what is actually being done.
And as luck would have it — thanks to AI — things are being done, and there may in fact be some room for optimism this year. A recent report by Google and Boston Consulting Group shows that AI has the potential to help mitigate 5-10% of global greenhouse gas (GHG) emissions by 2030 — the equivalent of the total annual emissions of the European Union.
Which is why it is worthwhile to discuss just how the technology of 2023 — Generative AI — can fit into the aims of COP28. In doing so, we can shed light on how it can not only be used by enterprises seeking to reduce their scope 1, 2 and 3 emissions, but also help humanity as a whole reach our net zero goals. Here are just a few of the known ways Generative AI can help combat climate change.
By simulating and evaluating various designs for solar and wind power systems, Generative AI enables the exploration of novel renewable energy configurations through the analysis of geographical and environmental data to tailor designs specifically to different locations and conditions.
In cases where real-world data is scarce or difficult to collect, Generative AI can also create synthetic datasets. General adversarial networks (GANs) can then be used to generate synthetic subsurface models, with the generator network of the GAN could be trained to produce synthetic models that are similar to real-world subsurface reservoirs, while the discriminator network would be trained to distinguish between real and synthetic reservoir models.
Once the generative model is trained, it can be used to generate a large number of synthetic reservoir models that could be used for reservoir simulation and optimisation, reducing uncertainty and improving hydrocarbon production forecasting.
In fact, Generative AI’s ability to forecast is already proving to be crucial in combating climate change. Google Research teamed up earlier this year with American Airlines and Breakthrough Energy to use AI to develop contrail forecast maps to test if pilots can choose routes that avoid creating contrails. After these test flights, Google found that the pilots reduced contrails by 54%. Contrails account for roughly 35% of aviation's global warming impact — which is over half the impact of the world’s jet fuel.
Outside of contrails, Generative AI can significantly enhance the accuracy of forecasting renewable energy production. Generative AI creates complex models that consider a wide range of environmental factors, crucial for predicting the output of unpredictable renewable sources like solar and wind energy. In the same vein, GenAI can simulate intricate and dynamic climate systems by generating comprehensive models that account for various environmental, geological, and atmospheric variables, helping us prepare for significant weather events via catastrophe modelling.
In terms of grid integration, Generative AI can help the integration of renewable energy sources by simulating different scenarios involving renewable energy inputs. In doing so, it helps with the planning and managing energy flow, storage, and distribution, ensuring that the transition to renewable sources maintains grid stability and efficiency.
On top of this, Generative AI can be used generate simulations of various carbon sequestration scenarios. This helps in understanding how different techniques might perform under various geological and environmental conditions, leading to the optimisation of carbon sequestration strategies. The technology can also simulate complex environmental systems to understand how different carbon reduction strategies might impact ecosystems. This is crucial in ensuring that carbon capture and reduction efforts do not inadvertently cause harm to the environment.
Generative AI can be used to design new materials or chemical compounds that could be more effective at capturing carbon. This involves simulating molecular structures and predicting their interaction with CO2, thereby aiding in the discovery of new materials for carbon capture. By simulating different materials and chemical compositions, GenAI can contribute to the development of more efficient and higher-capacity energy storage systems — crucial for managing the intermittency of renewable energy sources.
In fact, we saw this very recently with Google Deepmind’s announcement that their AI tool, Graph Networks for Materials Exploration (GNoME). GNomE identified an astonishing 2.2 million new crystals, including 380,000 stable materials with transformative potential for future technologies.
GNoME uses deep learning, particularly graph neural networks (GNN), to analyse and produce information about potential new materials, thus generating new knowledge in the field of material science as a result. It does this by predicting the stability of new materials and generating new crystal structures, enabling the development of sustainable solutions ranging from advanced batteries to efficient superconductors.
These are but a few of the ways in which Generative AI holds immense promise in aiding climate action. As with any scalable technology these opportunities must be enacted with heightened consideration due to their exponential potential and imbued with ethical considerations from the ground up in order to mitigate potential externalities. Successful and ethical implementation necessitates careful consideration of data ethics, bias mitigation, transparency, collaborative frameworks, regulatory adherence, and a climate-conscious approach to cooling of data centres in order to manage the environmental impact of AI.
However, notwithstanding the risks and the (ever-reducing) computing costs, the opportunities provided by Generative AI in terms of supply chain optimisation, novel material research and advanced forecasting means that the technology will be a key player in the climate emergency. Let’s just hope it has a seat at the table at this year’s COP.
If you would like to learn how to build a Generative AI strategy that is geared towards sustainability, get in touch today and we can help you build solutions that are the forefront of utilities and energy.