The fact is that most of these rogue antivirus programs all come from "families" - in other words, the malware authors actually design a development "kit" where any common criminal can create a rogue antivirus variant without any programming knowledge required. That's why you see so many of these rogues that look almost identical but have slightly differing names and graphics.
I imagine it's the same case with MacDefender. The question will be: does Apple's detection solution detect anything coming from the entire family of rogues, or just the individual rogues?
If it's the latter, then it's a very ineffective solution, as the malware authors only have to swap a few things around to create a new, undetectable variant, and Apple will have to play a game of catch up to stay on top of things. It could easily become unsustainable.
On the other hand, if Apple's solution is robust and can detect anything from an entire family of rogues, the pressure could be on the malware authors. They'd have to re-engineer a large portion of the code, and it would be trivial for Apple to render all of their hard work useless with a quiet definitions update within a few days. So then it becomes a cost vs benefit battle, hopefully with the malware authors giving up on their Mac test run and going back to the more lucrative Windows targets.