In This Story
Apple on Tuesday announced its latest iPad Pro with a new chip that it promises will majorly boost the efficiency of AI applications run on the device.
Apple’s new M4 chip includes a CPU (central processing unit), GPU (graphics processing unit), and an NPU (neural processing unit). CPUs are the brain of a computer, while GPUs are designed to boost the quality of computer graphics. NPUs are made specifically for machine learning and generative artificial intelligence tasks.
During the company’s “Let Loose” iPad event, Apple VP of hardware technologies Tim Millet said the M4 is an “outrageously powerful chip for AI” with its NPU “more powerful than any neural processing unit in any AI PC today.” Apple says the M4’s neural engine can perform 38 trillion operations per second, making it more than 60 times faster than its first generation NPU, the A11 Bionic chip. It also uses a lot less energy.
“Compared to M2 [Apple’s chip used in older generation iPads and MacBooks], M4 can deliver the same performance using just half the power,” Apple said in its announcement, “and compared to the latest PC chip in a thin and light laptop, M4 can deliver the same performance using just a quarter of the power.”
The announcement comes just less than two week before Microsoft is expected a announce a new Arm-powered chip that the company said would put its Windows laptops ahead of Apple’s MacBooks in terms of AI performance. Now, Microsoft could instead fall further behind.
Apple is not only developing chips to sit within its devices, but also chips to run artificial intelligence software from data centers, according to reports from The Wall Street Journal this week. The company is reportedly developing the AI chips in partnership with Taiwan Semiconductor Manufacturing Co. There’s no known timeline for the initiative code-named Project ACDC, for “Apple Chips in Data Center,” so it’s unclear when Apple would release such chips, if ever.