“The Energy Transition Won’t Happen”: Big Tech Embraces Energy Guzzling AI by Eric Worrall
Reality Sandwich with fried onions: The driving ambition to stay on top appears to have completely overwhelmed former tech company commitments to green virtue signalling.
“The Energy Transition Won’t Happen”: Big Tech Embraces Energy Guzzling AI
Essay by Eric Worrall
The driving ambition to stay on top appears to have completely overwhelmed former tech company commitments to green virtue signalling.
The “Energy Transition” Won’t Happen
Foundational innovation in cloud technology and artificial intelligence will require more energy than ever before—shattering any illusion that we will restrict supplies.
Mark P. Mills / Eye on the News / Infrastructure and energy, Economy, finance, and budgets, Technology and Innovation
May 23 2024The laptop class has rediscovered a basic truth: foundational innovation, once adoption proceeds at scale, is followed by an epic increase in energy consumption. It’s an iron law of our universe.
To illustrate that law, consider three recent examples, all vectors leading to the “shocking” discovery of radical increases in expected electricity demand, now occupying headlines today. First, there’s the electric car, which, if there were one in every garage, as enthusiasts hope, would roughly double residential neighborhood electricity demands. Next, there’s the idea of repatriating manufacturing, especially for semiconductors. This is arguably a “foundational innovation,” since policymakers are suddenly showing concern over the decades-long exit of such industries from the U.S. Restoring American manufacturing to, say, the global market share of just two decades ago would see industrial electricity demand soar by 50 percent.
And now the scions of software are discovering that both virtual reality and artificial intelligence, which emerge from the ineluctable mathematics of machine-learning algorithms, are anchored in the hard reality that everything uses energy. This is especially true for the blazing-fast and power-hungry chips that make AI possible. Nvidia, the leader of the AI-chip revolution and a Wall Street darling, has over the past three years alone shipped some 5 million high-power AI chips. To put this in perspective, every such AI chip uses roughly as much electricity each year as do three electric vehicles. And while the market appetite for electric vehicles is sagging and ultimately limited, the appetite for AI chips is explosive and essentially unlimited.
Consider a recent headline in the Wall Street Journal: “Big Tech’s Latest Obsession Is Finding Enough Energy”—because the “AI boom is fueling an insatiable appetite for electricity.” And, as Reuters reports, “U.S. electric utilities predict a tidal wave of new demand . . . . Nine of the top 10 U.S. electric utilities said data centers were a main source of customer growth.” Today’s forecasts see near-term growth in demand for electric power three times as great as in recent years. Rediscovery of the iron law of growth inspired an urgent Senate hearing on May 21 entitled “Opportunities, Risks, and Challenges Associated with Growth in Demand for Electric Power in the United States.” (Full disclosure; a hearing at which I testified.)
…
Each datacenter—and tens of thousands of them exist—has an energy appetite often greater than skyscrapers the size of the Empire State Building. And the nearly 1,000 so-called hyperscale datacenters each consume more energy than a steel mill (and this is before counting the impacts of piling on AI chips). The incredible level of power use derives directly from the fact that just ten square feet of a datacenter today has more computing horsepower than all the world’s computers circa 1980. And each square foot creates electric power demands 100 times greater than a square foot of a skyscraper. Even before the AI revolution, the world was adding tens of millions more square feet of datacenters each year.
…
Read more: https://www.city-journal.org/article/the-energy-transition-wont-happen
Bloomberg has a similar take on the situation;
Microsoft Wanted to be Carbon Negative. Then It Went Big on AI
If Microsoft is shooting for the moon on decarbonization, the moon is “more than five times as far away as it was in 2020,” President Brad Smith says on Zero.
By Akshat Rathi, Dina Bass, and Mythili Rao
23 May 2024 at 2:05 pm AESTIn 2020, Microsoft made an ambitious pledge: It would be carbon-negative by the end of the decade. The company was acting in response to the Intergovernmental Panel on Climate Change’s 2018 warning that, if the world wanted to keep warming below 1.5C, emissions needed to fall fast.
It was an ambitious target, but at the time it seemed achievable. Now that’s changed. New figures show that Microsoft’s total emissions in 2023 were about 30% higher than in 2020.
The confounding factor, Microsoft President Brad Smith tells Zero host Akshat Rathi, is artificial intelligence. After investing $1 billion in OpenAI in 2019, Microsoft has expanded its AI strategy — and its emissions have ballooned along the way.
The company says it isn’t abandoning its green goals, and has broadened its approach for reaching them. “You’ve gotta be willing to invest and pay for it, you’ve gotta be willing to persist,” Smith said. But can the tech giant have it both ways? Rathi sat down with Zero producer Mythili Rao to talk about his interview with Smith, and why Microsoft says the good AI can do for the world will outweigh its environmental impact.
…
This is the death of the green energy revolution we have all been waiting for. After 2020, tech companies have come to a shock realisation they need vast supplies of cheap energy, and fast. Tech companies still claim green energy is important to them, but personally I don’t believe these claims. Tech companies are clearly preparing to downgrade green energy in favour of more energy.
AI requires crazy amounts of power. All AI involves testing millions, sometimes billions of potential solutions every second of every day, to find the right response to stimuli.
Ever seen a wind blown leaf skitter across a pavement, and for just a moment your mind tricked you into thinking you saw a big spider? Or see a coiled rope or hose and thought you saw a snake? That illusion is a rare glimpse into how your brain works. Your brain continuously tests billions of possible solutions to interpret the flood of information being received from your senses – and occasionally gets it wrong, before rapidly correcting the mistake as more data becomes available.
AIs are no different, they also continuously test billions of possible solutions to find the solution which best fits the input they receive.
As AIs become more sophisticated, and demands grow for fewer mistakes, the sophistication and depth of the potential solutions being tested and discarded will rise exponentially. The energy required to run computers to calculate all these potential solutions will be unimaginable in today’s terms.
Just look at how bitcoin mining nowadays consumes more energy than some small countries, with millions of computers across the world testing and discarding uncounted billions of potential solutions to Bitcoin cryptography puzzles, to find that one solution which earns newly minted bitcoins for their operators. Bitcoin mining is just a taste of the energy expenditure the AI revolution will demand.
There is no path to green energy delivering the required magnitude of power, on anything like the timeframe tech companies need to stay competitive.
Tech companies know they need access to a vast increase in energy generation, and they need it fast, otherwise energy rich Asian AI companies will wipe the floor with them.
Media companies are desperate for front row access to artificial intelligence, to cut newsroom costs and prevent their rivals in India and China from using AI to identify which stories will draw the most audience attention, to identify tomorrow’s big news stories before anyone else.
Financial giants in New York, Chicago and San Francisco also have skyrocketing AI needs, to power their automated trading systems, and help them maintain their income, to help them keep winning in the cutthroat financial games they play against AI augmented rivals in Asia and Europe.
The funniest part of this tech company backflip is all the lefty companies which helped Biden win are likely now quietly in President Trump’s corner.
This surge in demand for energy to drive AI technology is very new, it likely took most of them by surprise. from the Bloomberg article quoted above, as recently as 2020 people in charge of big US tech companies were oblivious to how their energy needs would suddenly skyrocket.
I wonder what phone calls are quietly being made in places like New York, San Francisco and Chicago? Are obstacles to Trump 2024 quietly being dismantled, by people making backroom deals, former Democrat supporters who have come to the shock realisation that they need access to cheap energy more than they need green virtue signalling?
“Drill, baby, drill”
*Sniff Sniff*