WSJ: Apple trying to roll its own AI server chips

"The component will likely be focused on running AI models—what’s known as inference—rather than on training AI models."

For more, please login or subscribe