To switch or not to switch? Clean energy alternatives demand more energy as AI fuels unsustainable mining practices
A new debate has emerged about technologies that have become popular as the preferred pathways towards enabling countries to enhance climate change mitigation initiatives.
It is emerging that
these technologies, particularly electrification using renewable energy sources,
and the expanding influence of Artificial Intelligence (AI) pose serious threats
to the same environment they are seeking to protect. AI data centres have
become some of the largest consumers of electric power, potentially surpassing
global manufacturing.
A new report by
the International Energy Agency (IEA) has predicted that the annual energy
demand by AI data centres will be slightly more than that which is used by
Japan over the same period by 2030. Data centre’s demand for energy is
principally driven by the need for cooling of servers and other computing accessories. Due to increasing demand for
tools such as Chat GPT, data servers are increasing their data processing capacity and hence
requiring more energy to cool them down. In the US for example, the AI
industry’s electricity consumption is already projected to surpass electric
power demand by manufacturing by 2030.
The World finds
itself in catch 22-situation as a result.
Most of us have
come to accept that the use of renewable energy in place of fossil fuels and
tapping into AI for more efficient production confer benefit to efforts combat
the harmful effects of climate change. Whereas the logic of using renewable
energy is clear to most, hidden challenges relate to mobility that is powered by
electricity generated from renewable sources, one of the best examples being Electric
Vehicles (EVs).
What may not be
obvious to most is that batteries for storing energy for propelling EVs are made
of materials generally referred to as Rare Earth Elements (REEs). The mining of
these materials is highly complex. Also, many of them can only be found in places
where radioactive minerals such as Uranium and Thorium are in abundance, posing
a serious health risk to local populations due to the threat of radiation contamination.
Additionally, processes for extracting the materials require the use harmful
chemicals and large amount of water, which raises the risk of polluting ground water sources, an additional threat to public health.
Geopolitics of REEs
The other contentious
issues revolve around geopolitics. Today, China controls more than 70 per cent
of global production of REEs. It is the only country with a fully autonomous REE
value chain. Concern by Western countries about this situation stems from the
fact that China has the capacity to disrupt global supply chains for these
materials driving up costs and creating shortages that can paralyse the entire
renewable energy sector in the rest of the world.
The current
dilemma around AI is that the technology has become a key catalyst for
innovation, including those related to energy efficiency. The world is also
relying on AI-driven innovations to come up with battery technologies less dependent
on REEs, which will lead to reduced pressure to mine them, hence reducing the
resultant environmental damage. In the meantime, many countries are staring at
the prospect of reverting to recently abandoned use of unsustainable modes of power generation such as the use of petroleum products and coal, that are drivers of Greenhouse Gas (GHGs) emissions that cause global warming.
The IEA Report is quoted[1] as noting that countries that want to benefit from the potential of AI need to quickly accelerate new investments in electricity generation and grids, improve the efficiency and flexibility of data centres, and strengthen the dialogue between policy makers, the tech sector and the energy industry. It is only through figuring how all the increasingly complex intervening factors can be addressed so that global aspirations such as those relating to Net Zero targets will remain feasible.
The race towards automation, which is partially fueled by AI, is pointing to a situation where communities that will fall behind may face another new divide between the AI “haves” and the “have nots,” something that needs to also take the frontline as technology increasingly takes over many areas of human endeavour.
Comments