Microsoft and OpenAI announce that Stargate, a new AI super-computer data center project to be headquartered in the U.S. could cost over $115 billion and is planned for launch in 2028.
Sky rocketing demands for generative artificial intelligence have accelerated demands for AI-centric data centers that can handle more advanced tasks than traditional data centers.
The Stargate supercomputer project is designed over five phases. Expenses for the Stargate plan is more than triple Microsoft’s expenditure last year on capital spending for servers, buildings, and other equipment. With this capital outlay, the project would be over 100 times more expensive than some of the largest data centers operating today.
In addition, Stargate’s power requirements, estimated to be several gigawatts would likely require Microsoft and OpenAI to explore alternative power sources, like nuclear power.
One can only imagine the security implications of Stargate, especially if nuclear power is the primary power source.
Nuclear energy produces radioactive waste such as uranium mill tailings, spend (used reactor fuel), and other radioactive wastes. These materials can remain radioactive and dangerous to human health for thousands of years.
Nuclear waste is handled in compliance with the stringent requirements of the U.S. Nuclear Regulatory Commission, the U.S. Department of Energy, and the U.S. Environmental Protection Agency. From the time nuclear fuel leaves the reactor to when it waits in dry casks for permanent disposal, the U.S. nuclear industry sets an impressive standard.
What is different, however, with Stargate is the volume of nuclear waste will dramatically increase. Imagine we have hundreds of Stargate’s operating globally, as this is the direction we are heading.
Ask yourself the ethical question, what is the impact if nuclear waste volume explodes 100-fold?
Microsoft and OpenAI need to do robust scenario planning and work with leading scientists to understand the risks of not just one Stargate operating, but of hundreds of Stargate(s) or look alikes, and evaluate the impacts to Mother Earth.
There still remains the urgent requirement for increased governmental policy and legislation needed on generative AI risks.
We have also seen a lot of short-term thinking from OpenAI and numerous legal cases are in motion on copyright infringement in the USA, and around the globe.
Microsoft known for its ethical practices has indeed a tiger by the tail with its close relationship with OpenAI.
However, the recent investment of $2.1 billion in the French start-up Mistral AI buffers Microsoft a wee bit if the OpenAI relationship goes south with the ever-mounting court cases against OpenAI practices.
Read the full article here
