Loop Capital analyst Ananda Baruah reported on March 24 local time that Apple is ordering NVIDIA's GB300 NVL72 systems, with the order valued at approximately US$1 billion, equivalent to about 250 servers, each priced between US$3.7 million and US$4 million. "Apple has officially joined the competition for large AI server clusters," with the company collaborating with Dell and Advanced Micro Devices (AMD) to develop these clusters to support generative AI applications.
Baruah indicated that Apple's strategic shift stems from challenges faced by its AI-powered Siri digital assistant during its market rollout. The launch of the new Siri has been indefinitely postponed. Last June, Apple showcased Siri's AI capabilities at the Worldwide Developers Conference, initially hoping to release it early this year.
Reports suggest that due to difficulties in deploying its AI features, Apple has made adjustments to its executive team. One executive described these delays and missteps as ugly and embarrassing, especially given Apple's previous promotion of its AI capabilities in television advertisements.
Some analysts believe that Apple's procurement of NVIDIA AI servers may indicate that its in-house chip capabilities are insufficient to meet the growing demands of generative AI. As AI technology continues to evolve and its application scenarios expand, the demand for computational power is increasing sharply. Apple may need to seek more robust computational support while ensuring privacy protection.
Related News:
US Commerce Department to ban DeepSeek on government-furnished equipment
Comment