埃塞克斯價值 20 億英鎊的人工智慧超級電腦：為英國未來的科技革命提供動力！

Essex is set to become a major hub for artificial intelligence in the UK, thanks to the construction of a new £2 billion AI data centre in Loughton. This project, which is already underway, is expected to open in 2026 and will be the largest AI data centre in the country. Recently, Microsoft announced that it would invest in the site, helping to fund a supercomputer that will make the centre even more powerful. But what exactly is a supercomputer, and what will it be used for?

The Loughton data centre is a big deal for the region and for the UK as a whole. Microsoft’s involvement means the centre will have access to a share of a £22 billion investment pot, making it possible to build a supercomputer with thousands of devices working together around the clock. Nscale, the company behind the project, is teaming up with Microsoft and NVIDIA, which will supply the AI chips needed for the centre. When it opens, the centre will create up to 250 permanent jobs and is expected to support important services like healthcare and defence. The facility will cover four acres and will be filled with advanced computer systems designed to handle huge amounts of data.

The supercomputer at Loughton will be powered by 23,040 graphics processing units, or GPUs, which are specialized chips used to train and run AI models. At first, the centre will deliver 50 megawatts of AI capacity, but it is designed to grow and could eventually handle up to 90 megawatts. To put this in perspective, the world’s fastest supercomputer in the US uses about 30 megawatts at its peak, so Loughton’s centre could be nearly three times as powerful. These high-performance systems will be used to train AI models like ChatGPT and run real-time services such as voice assistants. With so many GPUs working together, the Essex supercomputer will be one of the most powerful AI computing centres ever built.

However, building and running such a large AI centre comes with challenges. One major issue is the amount of electricity needed to power the GPUs and keep them cool. If the computers overheat, they can be damaged, so cooling is very important. Energy prices in the UK are high, and there is pressure to use renewable energy sources to reduce environmental impact. For example, Google’s greenhouse gas emissions increased by 48 percent between 2019 and 2023, mainly because of its AI centres. Dr Haider Raza, an AI expert, says that high electricity costs could slow down the growth of data centres. The government is looking for clean energy solutions to support its AI ambitions. Another challenge is making sure the centre has strong internet connectivity for fast data transfer and robust cybersecurity to protect sensitive information. Prime Minister Sir Keir Starmer recently said that the UK is on its way to becoming a world leader in AI, which will mean more jobs, more investment, and better public services for everyone.