We can successfully predict some aspects of colony behavior using models that include only individual ants, without any global colony variables, showing that we understand how those colony behaviors arise from ant behaviors.

But that is just what is meant to be conveyed by the claim that intelligence emerges from the interaction of neurons. Of course that is trivially true. But, the original AI theory was that neurons were the building blocks of a universal computer (and how the universal computer was built of neurons wasn't particularly interesting). The problem of understanding intelligence was, therefore, the problem of understanding how the mind could be replicated in LISP code (or PROLOG or rule-based systems or etc).

The newer argument was that neurons have particular properties that have to be taken into account rather than ignored when attempting to understand and model intelligent behavior. And this is what is meant by saying intelligence 'emerges' from the interaction of neurons and is an 'emergent phenomenon'.