Intel has seen the light that AMD saw years ago; research breakthroughs fueling its innovation pipeline for keeping Moore's Law on track to a trillion transistors on a package in the next decade.... Intel Research Predict Trillion Transistors by 2030 (by using chiplets)
Gluing CPUs together was out of the question during the reign of the previous Intel CEO. No wonder he was fired. Considering everything Intel has achieved during its long existence, it would be pretty weird to categorically reject a viable, useful, and proven technological innovation.
It really has little to do with CEOs, it's just marketing. When the competition is offering something you don't have an answer to yet, you attack the solution instead of the product itself and use ad populum. You then drop the attacks just before you release your matching solution, and insist on calling your version something other than what everyone already calls it. (tiles vs chiplets) Companies do this all the time regardless of who is in charge.
Regardless, Intel kicked out the CEO of that period and, by the looks of it, hired one that doesn't look down on the MCM technology. So, maybe it wasn't only empty marketing material but the message was also ill received within the company?
This^^ And we can`t forget that Intel was the first company to use glue and AMD also made fun of it at the time.
Not to mention, cost savings. We all know how much Intel prioritizes profit margins. Considering their sub-par yield with 10nm, seems to me making smaller dies would have helped a whole lot with that.