Michael Burry just laid out a long case for $META arguing that hyperscaler AI spending could lift its implied AI-services margins by ~10%.
Pretty interesting coming from someone who's been so critical of AI spending boom and is now pointing to where those investments could actually pay off.