Introduction to Data Centers
If you ever fly to Washington DC, look out the window as you land at Dulles Airport – and you might catch a glimpse of today’s biggest economic story. Below you will see a series of huge warehouses scattered across the surrounding fields and forests, which to the untrained eye might look like supermarkets or distribution centers. But no: they are actually data centers – the largest concentration of data centers in the world.
The Role of Data Centers in AI
That’s because the area around Dulles Airport has more of these buildings, which house computer servers that do the calculations to train and run artificial intelligence (AI), than anywhere else. And with AI accounting for most of the U.S. economic growth so far this year, that makes this place a huge business. Down on the ground floor you can see the trademarks as you drive through the so-called “Rechenzentrumsgasse”. Huge power lines run everywhere, indicating that running these facilities is an incredibly energy-intensive task.
Energy Consumption
This tiny area alone, Loudoun County, uses about 4.9 gigawatts of electricity – more than the entire consumption of Denmark. This number has already tripled in the last six years and is expected to rise even further in the coming years. The Digital Dulles site, currently under construction, is expected to consume up to a gigawatt of electricity in total, with six substations expected to help provide that power. In fact, it uses about as much electricity as a large nuclear power plant.
Infrastructure and Limitations
Walking through the site, a series of large warehouses, some already equipped with rows of backup generators to ensure that the silicon chips whirring within them never lose power, is an impressive experience – a reminder of the physical foundations of the AI age. As weightless as this technology feels, it places enormous physical demands. This includes building these massive concrete buildings, each of which requires enormous amounts of electricity and water to cool the servers. The availability of this infrastructure is one of the main limiting factors for this economic boom in the coming years.
Economic Implications
Economist Jason Furman says the U.S. economy barely grew in the first half of this year, excluding AI and related technologies. So much depends on that. But there are some who question whether the U.S. will be able to build power plants quickly enough to fuel this boom. American electricity consumption remained more or less the same for years. That has changed rapidly in recent years. Now AI companies have made big promises about future computing power, but that depends on whether these chips can be connected to the grid.
The Risk of a Financial Bubble
Last week, International Monetary Fund chief economist Pierre-Olivier Gourinchas warned that AI could actually be a financial bubble. He said: "There are echoes of the current surge in technology investment from the dot-com boom of the late 1990s. Then it was the internet… now it’s AI. We’re seeing rising valuations, booming investment and strong consumption on the back of solid capital gains. The risk is that with stronger investment and consumption, tighter monetary policy will be required to contain price pressures. This has happened Late 1990s.”
Uncertainty and Future Growth
For those in the AI world, this also feels like uncharted territory. Helen Toner, executive director of Georgetown’s Center for Security and Emerging Technology, said: "The scary thing is, no one knows how far AI will come, and no one really knows how much economic growth will come from it." "The trend is certainly for the AI systems we’re developing to become more and more sophisticated over time, and I don’t see any sign of that stopping. I think they will become more and more advanced. But the question is, how much productivity growth will that create? How will that compare to the absolutely staggering investments that are being made today?" Whether it’s a new industrial revolution or a bubble – or both – there’s no denying that AI is a massive economic story with huge implications. For energy. For materials. For jobs. We just don’t know how massive yet.
