AI driven energy demand surge sees humans compete with bots
While AI talks an exciting game, it also has an insatiable appetite for energy, leaving thousands of people wondering where they’ll get their electricity from once even more AI data centres are built
READING LEVEL: ORANGE
We all know the AI revolution is here but most of us are a lot more in the dark about the power problem it’s rapidly generating.
All that computing genius requires an immense amount of electricity and water – both of which are in short supply for many communities across the world. And as climate change threatens our ability to safeguard these vital resources, the question remains: is AI worth its weight in resourcing gold?
THE AI ECONOMY
Tech CEOs have long promised a glossy future where machines will do all the boring legwork for us and help us solve the many problems of the world.
The advances achieved so far in this relatively new field have been exciting – and they have made a lot of money for the businesses that launched them.
In fact, AI has done so well that it has almost single-handedly carried the post-Covid* US economy, with big tech spending even more on further development.
Amazon, Microsoft, Alphabet and Meta alone are projected to spend roughly US$630 billion (A$879 billion) on data centres* and AI chips* in 2026.
Consulting firm McKinsey estimates global spending on data centres could hit US$7 trillion (A$9.77 trillion) by 2030, with the US accounting for a massive share.
For most of us, it’s all just some distant digital contest happening behind computer screens.
But the extensive infrastructure* needed to power that boom, captained by a tiny group of billionaires, is very real.
POWERING AI
Tens of thousands of massive data centres power AI across the world – with more being constructed each year. Most Australians won’t notice these facilities, but the scale behind them is still enormous.
The centres operate around the clock, consuming vast amounts of electricity while relying on massive cooling systems to stop servers overheating. Some forecasts suggest Australia’s data-centre capacity could more than double by 2030 as governments and corporations invest in AI infrastructure.
In fact, Sydney and Melbourne have been rapidly emerging as major AI and cloud-computing* hubs for the Asia-Pacific region. Amazon recently announced plans to invest roughly $20 billion into Australian AI infrastructure over five years, while Microsoft, Google and local operators such as NextDC continue expanding their data-centre footprint.
The Australian Energy Market Commission has warned rapid expansion of large-scale data facilities could create risks to the country’s stable supply of electricity if there are too many spikes in demand.
Data centres also consume huge amounts of water – one of our most vital resources for survival.
Overseas, researchers estimate global AI systems could consume more water annually than entire countries by next year.
While the extreme power and water requirements of AI have fuelled scepticism* and worry about the technology, governments see enormous economic potential in becoming regional AI leaders.
Technology companies argue the industry will drive productivity*, jobs and investment for decades to come. Several operators have also been investing heavily in renewable energy* to offset the immense power demands attached to the sector.
AI RESHAPING COMMUNITIES
As the AI race powers on, entire regions have started being quietly reshaped around the infrastructure required to power it.
In Ireland, data centres have been found to consume more than one-fifth of the country’s electricity supply. Parts of the Netherlands have pushed back against new developments because of fears local grids were becoming overloaded.
In Northern Virginia — the largest data-centre market on earth — residents have complained about industrial expansion*, giant transmission lines and rising pressure on energy systems.
POWER FOR THE PEOPLE OR FOR AI?
Tens of thousands of residents living at the popular tourist destination Lake Tahoe, which lies between the states of California and Nevada on the western side of the USA, have been hit with uncertainty over where their electricity will come from after 2027.
Utility provider NV Energy recently told the region’s supplier it would stop providing most of its power as demand surges elsewhere in Nevada due to data centres tied to the AI boom.
Several hours away from the picturesque lake, the northern Nevada city of Reno has experienced a tech boom, with big companies like Google, Microsoft and Apple expanding aggressively. Northern Nevada has transformed into one of America’s fastest-growing data-centre corridors.
Media publication Fortune has cited forecasts suggesting that proposed projects there could add nearly 5900 megawatts of electricity demand by 2033 — enough to rival the output of several major power stations.
NV Energy says the transition away from Tahoe had been planned for years and was not caused solely by data-centre growth. It said in a statement that it would continue to provide power to Lake Tahoe until the region’s supplier had found another source of energy.
But for locals, the uncertainty has become deeply personal.
“It’s like we don’t exist,” Danielle Hughes, a North Lake Tahoe resident and CEO of non-profit Tahoe Spark, told Fortune.
Hughes warned smaller communities risked being swallowed by the scale of the AI infrastructure race.
“We’re 49,000 customers. We have no leverage*,” she said. “We have no representation*. It’s resource extraction*.”
While AI’s potential has brought hope and wonder to many around what humanity could achieve in future, the situation down at Lake Tahoe has offered an uneasy glimpse into the trade-offs that could emerge as the technology expands further into the physical world.
POLL
GLOSSARY
- post-Covid: the main force driving the economy after the pandemic, when there was a great financial disruption
- data centres: facilities than house the IT infrastructure needed to train and run Artificial Intelligence (AI) applications and services, including servers, advanced storage systems, GPUs and networking equipment as well as cooling capabilities
- AI chips: specially designed computer microchips used in AI systems
- infrastructure: the buildings or physical structures that enable the delivery of a service or utility
- cloud-computing: the delivery of computing services over the internet
- scepticism: doubt as to the truth of something
- productivity: the rate of production by an economy
- renewable energy: energy from a source that isn’t depleted when used, such as wind or solar power
- industrial expansion: the growth of industry
- leverage: power to make a difference
- representation: lawyer to defend them
- resource extraction: taking away resources, such as energy
EXTRA READING
Trump’s AI alien after UFO claims
Today, AI changed Google forever
AI ‘cowgorithm’ to transform farms
QUICK QUIZ
1. By how much could Australia’s data centre capacity increase by 2030?
2. How much is Amazon investing in Australian AI infrastructure over the next five years?
3. What two important resources are required to run AI platforms across the globe?
4. Why are the residents of Lake Tahoe facing uncertainty over their future power supply?
5. How much of the country’s electricity supply do data centres consume in Ireland?
LISTEN TO THIS STORY
CLASSROOM ACTIVITIES
1. Follow the five Ws
Write down the key points from this Kids News article in short, succinct sentences:
Who:
What:
When:
Where:
Why:
How:
Possible headline for the article:
Time: allow 15 minutes to complete this activity
Curriculum Links: English, Critical and Creative Thinking
2. Extension
Write down the arguments for and against and points of view presented in this article.
FOR: AGAINST:
Time: allow 10 minutes to complete this activity
Curriculum Links: English, Critical and Creative Thinking
VCOP ACTIVITY
Wow word recycle
There are plenty of wow words (ambitious pieces of vocabulary) being used in the article. Some are in the glossary, but there might be extra ones from the article that you think are exceptional as well.
Identify all the words in the article that you think are not common words, and particularly good choices for the writer to have chosen.
Select three words you have highlighted to recycle into your own sentences.
If any of the words you identified are not in the glossary, write up your own glossary for them.
Extension
Find a bland sentence from the article to up-level. Can you add more detail and description? Can you replace any base words with more specific synonyms?
Down-level for a younger audience. Find a sentence in the article that is high level. Now rewrite it for a younger audience so they can understand the words without using the glossary.