OpenAI, Microsoft to invest $100 for world‘s largest data center

이미지 확대

[Courtesy of each company]

OpenAI Inc., the developer behind generative synthetic intelligence (AI) ChatGPT, plans to speculate $100 billion to construct the world’s largest knowledge middle in partnership with Microsoft Corp. The middle that’s envisioned might be outfitted with a supercomputer that might be utilized in growing synthetic common Intelligence (AGI). The funding alone is anticipated to be 100 occasions the dimensions of the present largest knowledge middle.

In keeping with info know-how (IT)-focused enterprise publication ‘The Info’ on Friday (native time), the administration of each corporations is engaged on a challenge named ‘Stargate’ to construct a knowledge middle outfitted with particular AI chips to energy OpenAI companies.

The core of the envisioned knowledge middle is a supercomputer, which can home hundreds of thousands of specifically designed server chips to energy OpenAI‘s AI fashions. Provided that Microsoft and OpenAI are reportedly planning a knowledge middle with an preliminary value of $100 billion, they might make important investments transferring ahead to make sure ample AI computing capability.

“That is a necessary step in constructing AGI,” Chris Sharp, chief know-how officer (CTO) at knowledge middle operator Digital Realty Belief Inc., mentioned. “(The funding) could appear unimaginably massive by present requirements, nevertheless it received’t seem so as soon as the supercomputer is definitely constructed.”

The Stargate challenge is a part of a five-stage AI infrastructure building challenge. OpenAI and Microsoft are presently halfway via stage three, and aiming to unveil the OpenAI supercomputer at stage 4 in 2026, which can value $10 billion. A good portion of the prices in levels 4 and 5 might be associated to the acquisition of AI chips., in accordance with the publication.

Former OpenAI CEO Sam Altman and and Microsoft CEO Satya Nadella at OpenAI’s DevDay in San Francisco on Nov. 6, 2023. [Photo by Yonhap]

See also  OpenAI data residency advances enterprise AI governance

Constructing a $100 billion knowledge middle would require hundreds of thousands of semiconductors and though Nvidia Corp.‘s AI chips, similar to graphic processing models (GPUs), are doubtless candidates, it’s also attainable that Microsoft will use its personal AI chips. There may be a surge in demand for high-bandwidth reminiscence (HBM) from SK hynix Inc. and Samsung Electronics Co., which might be used alongside GPUs. Demand for Intel or Arm-based CPUs for AI servers may even improve considerably.

“We’re at all times planning for the subsequent era of infrastructure innovation essential to increase AI capabilities,” a Microsoft spokesperson mentioned in response to a Reuters request for remark.

OpenAI additionally unveiled Voice Engine, an AI instrument that learns human speech and generates imitation voices, on Friday. OpenAI shared preliminary experimental outcomes of the Voice Engine on its weblog, which confirmed it may generate a voice just like the unique speaker utilizing a 15-second voice pattern.

Nevertheless, given the potential dangers of this AI instrument, the corporate will give cautious thought on whether or not to launch it in full.

By Lee Dong-in, Oh Chan-jong and Chang Iou-chung

[ⓒ Pulse by Maeil Business News Korea & mk.co.kr, All rights reserved]