Amidst concerns over the energy and water-intensive nature of training artificial intelligence (AI) programs, critics often overlook the crucial role AI will play in the green transition and combating climate change. It is essential to recognize the long-term benefits AI can bring and avoid succumbing to fear and apathy driven by misleading headlines.
—
We have seen a litany of headlines lambasting how energy and water-intensive it can be to train artificial intelligence (AI) programmes.
One analysis suggests that by 2027, the AI sector could consume between 85 and 134 terawatt hours each year. Similarly, AI usage demand may require up to 4.2-6.6 billion cubic metres of water withdrawal in 2027. For context, that is the same amount of energy and water consumed annually by the whole of the Netherlands.
This, of course, has triggered a wave of environmental concerns. Yet this criticism seems to highlight a blindspot in the “green movement”; the green transition will require energy investments, whether we like it or not.
More on the topic: The Real Environmental Impact of AI
We have seen this before; the energy intensity of creating and developing electric vehicles, wind turbines, and lab-grown meat have all come under similar criticism. What they always miss is the emissions that could be abated through the creation of these technologies.
While many of AI’s applications will not have anything to do with decarbonisation, some will.
Ultimately, climate change is a technical problem. As with any problem, to solve it, you must first understand it. AI will grant us a far better understanding of exactly what is causing climate change, and how. For example, scientists at the University of Leeds can already map large antarctic icebergs in just one-hundredth of a second. As data on climate change becomes readily available, our ability to deploy the right resources to the right places will only grow.
Secondly, AI will help us update our legacy systems. It will be crucial for optimising energy use in buildings, transportation, and manufacturing. Crucially it will help to balance an increasingly less resilient and reliable grid from increased renewables penetration by more accurately forecasting demands and predicting the output of renewable energy assets.
AI has already been indispensable, as we grapple with the challenge of feeding more people with fewer, failing crops. From monitoring crop health, micro-dosing pesticides, to predicting future weather patterns, AI will become synonymous with the farms of the future.
Enhanced supply chain management, real time environmental monitoring, accurate climate change analysis, renewable energy production; the list goes on.
Finally, AI will deliver those technologies that we need, but do not exist yet. Simply cutting emissions alone will not solve climate change, when we do not have appropriate technologies to fill the gap. For example, AI has already been used to accelerate the arrival of fusion technology, which may just be our ticket to near-infinite, clean energy.
Essentially, AI is implicit in every technology used in the green transition, to the extent that the transition cannot be made without it. That is why these headlines that lambast AI’s energy usage are so short-sighted.
While AI’s energy and water usage are cause for attention, we cannot allow that to overshadow the enormous benefits the technology will bring. Indeed, it is the one technology that allows our other technologies to reach their full potential.
All of this points to a fundamental error in the collective thinking on climate change. The notion that we already have all the technologies we need, without the need for any further innovation, is delusional. If the countries attempted to run on renewable energy alone, or go without fossil fuel based fertiliser, a global humanitarian catastrophe would quickly ensue.
To have any real chance of mitigating climate change, we must take a long-term, strategic view. Short-term emission reductions are essential. But we mustn’t cow in fear at scary-looking emission profiles of new technologies, while ignoring their enormous emission-abating potential.
Headlines like this tend to instil fear and its more dangerous cousin, apathy. In reality, AI is not the environmental scourge we are being told it is, it may well be our saviour.
