This is really a common question, and I believe I can sufficiently answer it.
Here's the TLDR: The MAME project will continue as long as there's sufficient interest and sufficient developers to keep things going.
MAME started life as a collection of Pac-Man emulators. It expanded out to include a few other games such as Mappy. It expanded out to include a ton of arcade machines.
MAME gave birth to MESS, which used the knowledge gained to work on computers, consoles, and handhelds. Eventually that was folded back into MAME itself as it made more sense to keep it together in a tight package so that improvements were more readily noticeable across the board.
MAME gave birth to PinMAME, and while the code couldn't be folded back in, a lot of knowledge did eventually make it back into MAME proper.
To condense all of that down, MAME has been a gradually increasing scope since the very first day. The number of targets for things we seek to preserve is never truly going to run low. Yet, many of the original MAME developers such as Nicola Salmoria himself have retired over the years.
As long as there's even a single developer interested in working on things, MAME will continue to move forward and preserve hardware at risk of being lost. It would just help things dramatically if we could catch the interest of additional volunteers to help work on areas that need further manpower.
Try checking the MAME manual at http://docs.mamedev.org