I am no Greta Thunberg, but Coldplay’s sustainability report shocked me for all the right reasons. The band made history by becoming the first ever to produce a sustainability report, aiming to reduce carbon emissions by 50%. They are achieving this through measures like reusable LED wristbands made from plant-based materials, which are 100% compostable. Additionally, the stages they use are made of lightweight, low-carbon material and reusable and recyclable steel. Greta Thunberg would be mighty proud. | Can Sustainability and Profitability Go Hand In Hand ?
If a pop band could do this, what is stopping other companies? The answer lies in the costs of making sustainability reports. IBM research found that the costs of sustainability reporting exceed the costs of sustainability innovation by 38% in India, highlighting a need for systems facilitating better data management and reporting.
In today’s world, sustainability is not optional but an unavoidable necessity. Surprisingly, the IBM report also found that organizations in India that embed sustainability are 41%more likely to attribute great improvement in revenue from their efforts and 90% more likely to outperform their peers in profitability. The good news is that 63% of executives surveyed believe that GenAI will play a crucial role in their sustainability efforts, with 76% of Indian organizations intending to boost investment in GenAI for this purpose.
Are you wondering how GenAI helps in achieving sustainability? Here’s how:-
Automation
AI at its core is designed to help humans do their routine tasks much more efficiently, with minimal supervision. The productivity benefits of AI applications in the four key sectors can generate a potential gain of US$3.6-5.2 trillion through optimized use of inputs, higher output productivity, and automation of manual tasks. Additionally, these applications can accelerate the move to a low-carbon world, reducing worldwide greenhouse gas emissions by 0.9-2.4 gigatons of CO2e.
Managing Supply Chains
AI-driven inventory management can help reduce excess inventory, optimize transportation, and enhance factory-supplier collaboration and alignment. GenAI can also aid in demand forecasting, and route optimization, minimizing waste and reducing the carbon footprint. Walmart has successfully adopted AI-powered supply chain management, which has resulted in substantial savings and an improved environmental impact.
Making Recycling More Efficient
AI can be used in recycling processes to automate sorting, prevent contamination, and predict maintenance needs, thus improving recycling efficiency and boosting sustainability.
Managing Power Used for IT Efforts
Organizations can now accurately forecast software requirements, real estate capacity usage, and data center capacity using AI. This allows an organization to better meet its sustainability target.
Monitoring & Controlling Energy
AI-powered systems can continuously analyze data from sensors and devices to optimize energy usage, water consumption, and resources thus helping you reduce waste and carbon footprints.
Is AI Sustainable?
AI is no doubt capable of helping you achieve your sustainable goals. But is the technology that has taken the world by storm for all its benefits to mankind sustainable in the first place?
Researchers argue that training a single large language deep learning model such as OpenAI’s GPT-4 or Google’s PaLM is estimated to use around 300 tons of CO2 - while an average person is responsible for creating around 5 tons of CO2 a year. Researchers at the University of Massachusetts found that training a single AI model can emit as much carbon as 5 cars in their lifetime.
How to Make AI Greener?
Use Existing Large Generative Models
There are already many providers of large language models, which are readily available in the market. These models are cheaper than making your own and more sustainable as they are already trained.
Fine-Tune Existing Models
Rather than training your own AI model, it is more environmentally friendly to refine an existing model. Fine-tuning and prompt training on specific content domains consume much less energy than training new large models from scratch.
Use Energy-Conserving Computational Methods
Using less computationally expensive approaches is another way to reduce energy consumption. The TinyML consumes a few hundred microwatts — a thousand times less power — to process the data locally without sending it to data servers.
We live in a tech-driven world, obsessed with transcending to newer possibilities. But in the process, we mustn’t lose sight of preserving the environment, and risk losing the possibility of life.
(Author:- Karishma is a content writer at Techdoquest. She gives youthful and refreshing perspectives on whatever she writes about.)