AI uses technology to solve problems, usually with huge amounts of data. The result is software that can have conversations, act as your virtual assistant and write essays, among many other things.
The problem is, processing enormous amounts of data at top speed uses a huge amount of energy, which of course has a knock-on effect on the environment.
The Good:
AI is helping to make lots of things more efficient. Doctors are using AI tools to diagnose diseases quickly and more accurately. Car manufacturers are developing self-driving cars, which could make our roads safer. And teachers are using AI to create exciting, engaging lesson plans.
AI is helping to boost profits and has already become a key part of future strategy for many companies. NVIDIA’s chips power so much of this AI infrastructure that in 2024, its stock price surged by over 170%. Demand for computing power has increased a lot, especially from sectors like gaming, data centres and AI research.
Microsoft and Alphabet have also had their stock prices rise as they’ve carried on integrating AI into their products and services. Microsoft’s Azure AI and AI-powered Office tools have boosted their strong earnings growth, pushing their stock price up by about 15%. Alphabet is investing heavily in AI and using it to improve Google’s search algorithms, cloud computing and autonomous driving technology. Alphabet’s stock rose by more than 35% over 2024.
The Bad:
AI makes processes more efficient, but it also comes with a few big risks.
One of the biggest concerns right now is how many jobs will be eliminated by AI. Many jobs currently done by humans have the potential to be automated, and AI has already started replacing us in manufacturing, logistics and customer service roles.
As AI gets smarter, even specialised professions like law and medicine could be automated, which could lead to high levels of unemployment.
As automation becomes more widespread, we also need to think about increased security risks. In the future AI could become part of critical infrastructure like our ambulance services, which could open these services up to dangerous cyberattacks. AI could also be used in the military, for example with autonomous weapons, which could have disastrous consequences.
Companies are also using AI to automate decision-making. It can help to filter potential job candidates when hiring, or help banks decide who to lend to. Unfortunately, automated decision making like this has potential for bias. It uses data for automatic profiling, which could potentially make unfair decisions and encourage discrimination based on the data sample it's trained with.
The Not So Eco-Friendly:
AI is changing the world, but it comes at an environmental cost. AI models use a significant amount of energy, as they need powerful graphic processing units and specialised chips to process massive data sets.
Training a large AI model like OpenAI’s GPT-3 could use as much energy as the average household uses over several hundred years. Plus, it’s estimated that every day, ChatGPT uses enough energy to charge 8 million phones.
Data centres also rely on water-intensive cooling systems to stop overheating, putting a huge strain on local water resources, especially in areas where water is already scarce.
Both water and electricity, or lack of it, raise big questions. AI is certainly useful, but whether it’s sustainable is another question entirely.
Striking a Balance
AI has huge potential. It could boost global efficiency and accelerate economic growth, but it’s undeniable that its environmental footprint is a big concern.
If companies do decide to use AI, they’ll also need to work hard on being sustainable to help reduce the impact AI has on the environment. For example, they could use renewable energy for data centres, optimise their algorithms to lower the amount of energy they use, or use more efficient water-cooling systems to make sure they’re not affecting the environment in a negative way.
Innovation by AI itself could also help address climate change. AI has already been used to test climate scenarios, optimise energy consumption in smart grids and improve farming with precision agriculture. If we use AI to help create sustainable practices, we can potentially lower its overall environmental impact.