Home Industry Big Data Big Tech plans to supply AI's ...
Big Data
Business Fortune
04 November, 2024
There is no indication that the massive global increase in data centers will slow down, which is forcing Big Tech to think about how to best support the AI revolution.
Among the possibilities is quantum computing, liquid cooling for data centers, and a shift to nuclear. However, some have argued that tech companies should abandon the "move fast and break things" mentality and acknowledge the cost of the generative AI boom throughout the whole supply chain as the rate of efficiency increases in power consumption slows.
The chief of the division, Somya Joshi, stated in a video call with CNBC that the true environmental cost is now very obscured. The necessity for IT companies to offer a product and support merely supports it.
The International Energy Agency says that growing digitalization and the use of generative AI will be the primary forces behind a surge in data center investment that is expected to accelerate further in the coming years.
The engineering group's data center sector has had impressive growth in recent years, with the category expected to rise by more than 24% in 2024, according to Giampiero Frisio, president of electrification at Swiss multinational ABB.
As they look to bring more energy capacity online to train and run the enormous generative AI models underlying today's apps, U.S. tech giants Microsoft, Google, and Amazon have all recently won nuclear energy for AI partnerships totaling billions of dollars.
The need for generative AI has increased along with the search for more efficient cooling solutions for data centers, particularly liquid cooling, which utilizes water to cool servers and other electrical equipment.
The CEO of Quantinuum, Raj Hazra, stated in a video call with CNBC that he believes that any brilliant invention that is discovered in the summer has a winter, but that you should ignore it till winter hits. He uses that to explain the state of generative AI, the infrastructure required to support it, and the enormous data centers that must be constructed.
Hopes for the generative AI boom are already putting pressure on the cost of running the technology, Hazra said.