AI and Environmental Sustainability – The Good and Bad

In the last few years, AI (Artificial Intelligence) has become one of the most widely proliferated new technologies for business since the advent of the internet.

What was once a science-fiction curiosity has quickly become an essential tool across several business sectors. AI’s potential applications are enormous, but so are the potential challenges it brings.

Sustainability was once an ethical choice, but it is now a strategic necessity for many businesses. Tightening regulations, ethical investors, consumer demands, and the effects of climate change are forcing businesses to consider their impact on the world through ESG (Environmental, Social and Governance).

So, how does AI fit into the sustainability puzzle for businesses? This blog will explore its benefits and disadvantages.

How AI can improve sustainability

To start, let’s examine some ways that AI can be a powerful tool to aid sustainability efforts. This list addresses issues in broad strokes, and it is important to exercise caution when applying it to specific businesses.

Identifying inefficiencies and cutting waste

AI tools can analyse enormous data sets, identify patterns, and provide decision-makers with actionable insights into efficiency. They can highlight inefficiencies in energy, water, and material consumption. For example, AI-powered smart grids can optimise energy distribution.

More efficient manufacturing

AI-driven predictive maintenance can reduce manufacturing downtime and resource waste in factories. It can also predict equipment failures long before they occur, optimise maintenance schedules and make things more efficient overall.

AI and the Circular Economy

AI could play a very important role in advancing the circular economy. It can do this by helping businesses optimise their recycling processes and manage product lifecycles better.

For example, companies may deploy AI to help them disassemble and recycle old devices with greater efficiency. Furthermore, AI helps track products throughout their lifecycle, from production to end-of-life, promoting recycling and waste reduction at each step.

AI and Smarter Agriculture

AI can improve precision farming practices. Farmers can use AI to make better use of water, fertilisers, and pesticides. AI can also help improve the precision of irrigation, monitor water quality, detect leaks, analyse soil quality, control fertiliser application rates, and detect pests and diseases.

Sustainability issues with AI

Critics of AI proliferation and environmental campaigning groups have expressed concerns about its environmental impact, with particular concern for carbon emissions, environmental waste and ecosystem damage. Here are some of the key sustainability issues linked to AI innovation.

High energy consumption of AI

The training of AI models uses large data centres, which consume enormous amounts of energy.

In 2019, the University of Massachusetts Amherst conducted a study revealing that training a single AI model could produce as much carbon dioxide as five cars over their lifetimes.

The iterative process of training and fine-tuning AI models is also resource-intensive, which may not be considered an environmentally responsible practice.

Proliferation of e-Waste

As new technologies develop rapidly, older technology becomes obsolete. This is true for AI, which is believed to drive faster hardware obsolescence and, therefore, an increase in electronic waste worldwide. Furthermore, many AI applications require specialist hardware, which can become outdated and end up in the scrap heap very quickly.

Increased natural resource consumption

The production of AI hardware necessitates increased rates of extracting and refining raw minerals, which is linked to environmental degradation, habitat destruction, and pollution.

The carbon footprint of AI

Running AI systems, especially in large-scale applications, results in substantial carbon emissions. This energy use includes electricity generation to power data centres, which are essential for running AI applications.

Data centres not only consume large amounts of electricity but also generate heat, necessitating additional energy for cooling systems. These factors contribute to a significant carbon footprint associated with AI systems in large-scale applications.

Habitat and biodiversity destruction

The construction and upkeep of the essential physical infrastructure for AI tools, which includes data centres and networking equipment, can have significant environmental impacts, such as altering land use, contributing to deforestation, and leading to the loss of biodiversity.

In addition, improper disposal of AI hardware can lead to toxic substances leaching into soil and water, causing environmental harm.

What AI means for your ESG priorities

AI can be a powerful tool for improving a company's ESG if used appropriately. However, it is important to carefully consider and implement strategies to address risks and ensure responsible development and usage.

As expectations for strong ESG practices in business grow, extra attention must be paid to the impact that the adoption of new technologies may have on a company’s environmental, social, and governance policies. For a lot of businesses, especially those seeking B-Corp status, AI is perhaps the area where this applies most in 2024.