World Earth Day (April 22) was originally just a day of reflection. The life-sustaining properties of this planet are unlike any other celestial body observed, and this day of contemplation is intended to remind us all that it is our solemn duty to protect all that the Earth has to offer us.
While there is still a lot of work to be done in the shift to renewable energy and a more sustainable lifestyle, some organizations are taking small steps towards building a better world by using modern AI tools. Here's an interesting story about how AI is being used to save the planet.
IBM
IBM has a track record of excellence in AI and data, so it makes sense for the company to leverage this expertise in the fight against climate change. IBM is dedicating its computing resources to a variety of environmentally-focused efforts.
First, IBM established IBM Global Real Estate to track, analyze and report on progress toward specific sustainability goals. Every year, the AI program collects data from more than 6,500 utility bills received around the world, aggregating total energy consumption, costs, and renewable energy purchases. This allows the company to generate reports and filter this information by region, geography and utility, allowing IBM to understand where energy savings are needed most.
In addition, IBM is working with organizations to accelerate the transition to low-carbon energy sources and change the way they currently operate to ensure they are more climate-resilient. IBM specifically mentioned its partnership with supermarket chain Sales Group to help balance electricity consumption associated with renewable energy supply. Using artificial intelligence and IoT sensors, IBM is providing the ability to make smart energy decisions. Large buildings, such as grocery stores, can partially suspend their energy use for heating or cooling until a certain threshold is determined that does not impact operations.
IBM is also working to help companies improve climate resilience by providing them with the tools they need to predict the impacts of climate change. The company is developing a geospatially based model that will be able to track deforestation, detect greenhouse gases, and even predict crop yields. As climate change makes our world more unpredictable than ever, these AI-powered tools will become increasingly important.
NASA
IBM's collaboration with the Foundation model in the field of environmental protection has had a profound impact. NASA is celebrating a partnership with the company to create an artificial intelligence geospatial foundation model. The model was trained on the vast amounts of data generated by NASA's Harmonized Landsat and Sentinel-2, and it provides the basis for much AI research focused on environmental issues.
This model has been successful to some extent. Since partnering with IBM in 2023, NASA has been able to use the model to detect fire consequences, map flooding, and classify crops and other land-use categories.
NASA has a wealth of data on many aspects of the Earth's environmental problems. However, scientific organizations need a lot of computing power and AI algorithm optimization to process this data to achieve their goals. Therefore, partnering with IBM to provide these much-needed resources is a perfect match.
Humanity owes its existence to historically consistent climate and environmental patterns. While anomalies do occur, most human civilizations exist in a predictable world. Sadly, even the most rigorous projections of a future of climate change can introduce a degree of uncertainty, and this unpredictability is most evident in the increase in extreme weather events.
Google is working to help humans prepare for changes in the weather system. The search engine giant has developed the Scalable Integrated Envelope Diffusion Sampler (SEEDS) model. Similar to large language models such as ChatGPT, SEEDS uses generative AI technology to generate multiple possible weather scenarios at a fraction of the cost of traditional physics-based predictive models currently in use.
Unlike these models, SEEDS is based on the denoising diffusion probabilistic model, which is a state-of-the-art generative AI approach created in part by Google Research. With just one or two forecasts from an operational numerical weather prediction system, SEEDS can generate a large set that produces credible real-world weather scenarios. By directly simulating the joint distribution of atmospheric states, SEEDS can realistically capture the correlation between geographic covariance and mean sea level pressure and the tropospheric median potential. Weather forecasters typically utilize these two closely related criteria to evaluate and validate their forecasts.
The climate crisis is one of the greatest obstacles to humanity's continued survival. However, by listening to scientists and using our most valuable modern tools, we can begin to build a sustainable world for future generations.