Compute: A New Measure of the World

Compute once meant calculation. Now it shapes work, art, and power. An invisible current running through the modern world’s every action.

Compute: A New Measure of the World
TV, from analogue to streaming.

I was listening to Azeem Azhar speak about the rise of compute, the planetary build-out of data centres, GPUs, and artificial intelligence. His argument was that we are not simply investing in AI; we are industrialising computation itself. The numbers were staggering, but what caught me wasn’t the scale, it was the word compute. What does it actually represent?

Quick takeaways

  • “Compute” has become a new kind of utility, the electricity of thought.
  • We use it to extend logic, scale cognition, and simulate perception.
  • Yet the human experience of all this remains strangely familiar.
  • The deeper question is not whether we progress, but whether we can stop.

Reflection

A small video I once saw came to mind: a first-year computer-science lecturer telling his students that the course was not about computers. The term, he said, pre-dated the machine; a “computer” was once a person who computed. In that sense, compute is less a thing than an activity, a verb that escaped its subject.

When we speak of growing compute, we are really speaking of externalising more and more of what the mind can do. Early computation was bureaucratic: it helped us count, order, and store. Today it speaks, paints, and reasons with us. The instrument has entered the register of experience.

It feels inevitable, as if we are drawn along by the gradient of possibility itself. We compute because we can, and once we can, we must. The process satisfies both our instrumental needs, efficiency, mastery, labour, and our existential ones: expression, curiosity, play. The system feeds itself; the appetite grows with every new capacity.

💡
In recent years, compute has also become a noun. Tech circles now speak of “how much compute” a system or region possesses, and thus meaning the total available capacity to perform calculations. It compresses a vast idea into a single word: the sum of all processing power, energy, and infrastructure that makes digital work possible. The shift from verb to noun mirrors how computation itself has turned from an action into a resource, as something to be stored, traded, and scaled.

Power and infrastructure

Yet behind this mental expansion lies a struggle for power that is anything but abstract. Compute is not evenly distributed; it pools where money, materials, and policy converge. The race to build new data centres and chip factories is also a race to secure influence over the world’s thinking infrastructure. Each processor carries a fragment of sovereignty.

Transistor count over time. It gives an indication of the total rise of compute.

That is why governments treat semiconductors as strategic assets, why export bans feel like acts of war, and why companies invest trillions in hardware that few of us ever see. Control over compute is control over simulation, design, and decision, over how the future is imagined and executed.

Europe, with its rules and reflections, stands between dependence and autonomy, between ethical clarity and technological hesitation.

Closing

Perhaps that is why it feels both thrilling and weightless. We call it progress, but it may simply be the unfolding of what was always latent in our tools, the steady electrification of thought and feeling.

Whether this is advancement or merely continuation is a question only hindsight can answer, and by then the next layer of compute will already be running.

💡
Raj Reddy once suggested a simple test for what AI ought to be in human life. Not a destiny, not a replacement, but a posture. In his view, intelligence at scale earns its place when it acts as a cognitive amplifier that helps us think more clearly and do more with the time we have; and as a kind of guardian angel that notices risks and protects what we value before we can. The machines extend our reach and guard our blind spots, but they never cease to be in service of us.

🔮 The demand for infinite compute
There’s a deeper shift behind the headlines
OpenAI Says It’ll Build Five More Huge Data Centers. Here’s Where
The ChatGPT maker is teaming with Oracle and SoftBank on the facilities, at an expected cost of more than $300 billion.

Where the compute lives.

From Silicon to Intelligence: Understanding the Hardware Behind AI
A short video about NPUs and TPUs led to a deeper look at the physical side of AI. From the Neural Engine in your iPhone to the massive processors powering data-centre models.
Infinite scale: The architecture behind the Azure AI superfactory - The Official Microsoft Blog
Today, we are unveiling the next Fairwater site of Azure AI datacenters in Atlanta, Georgia. This purpose-built datacenter is connected to our first Fairwater site in Wisconsin, prior generations of AI supercomputers and the broader Azure global datacenter footprint to create the world’s first planet-scale AI superfactory. By packing computing power more densely than ever…
The AI Pioneer With Provocative Plans for Humanity | Quanta Magazine
While some fret about technology’s social impacts, Raj Reddy still believes in the power of artificial intelligence to improve lives.