Apple is reportedly in the process of developing chips specifically designed to run artificial intelligence (AI) software in data centers, as revealed by sources familiar with the matter to the Wall Street Journal on Tuesday.
The project, known as ACDC (Apple Chips in Data Center), has been in development for several years but lacks a definitive timeline for completion, according to the report. Collaborating with Taiwan Semiconductor Manufacturing Co., Apple has been focusing on both the design and production of these chips, although it remains uncertain whether the efforts have yielded tangible results.
The primary focus of Apple’s server chip is expected to be on AI inference rather than AI model training. This decision aligns with industry expectations, with experts predicting that the training of AI models will continue to be dominated by established players like Nvidia.
In AI, inference is the crucial process through which trained machine learning models draw conclusions from new data. Other major tech companies, including Google, have also ventured into developing their own AI inference server chips to reduce reliance on external chip designers.
Despite its recent efforts, Apple has faced criticism for lagging behind competitors in the AI race. Analysts have noted significant investments from rivals like Microsoft in AI technology and infrastructure.
Apple’s CEO Tim Cook hinted at the company’s significant investment in AI technology during a statement in February, suggesting that an AI-related announcement could be expected later in the year, possibly during the Worldwide Developers Conference scheduled for June.
During Apple’s fiscal second-quarter earnings call, Tim Cook reiterated the company’s commitment to AI investment, emphasizing their belief in its transformative potential. Despite a 4% drop in overall revenue attributed to declining iPhone sales, Apple announced its largest-ever share buyback of $110 billion during the earnings call, resulting in a surge in share price.
