Cool. I'm going to guess that the ant was crawling along the walls or floor of the microwave (rather than on the food or plate). If this is true, then the ant must be smarter than we think. The ant must know that the electric field vanishes at the surface of a conductor due to the fact that charges (electrons) in the metal can immediately move in response to an incident electric field in such a way as to exactly cancel the field. This is the basis for Faraday cages and other related phenomena.
By being right next to the metal, the ant avoided the region of high intensity electromagnetic fields. Let's see... the wavelength of the microwaves in a microwave oven is about 12 cm, meaning the microwaves don't reach full intensity until about 3 cm off the deck (one-quarter wavelength). If the ant is only 2 mm high, the electric field amplitude is only about 10% its peak value at the location of the ant's back. But power density, it turns out, is proportional to the square of the electric field, and is therefore only 1% of the full power density away from the walls.
Depending on whether you want to trust my theory or not, you may try the following experiment. Put a coffee mug of water in the oven (though not part of the experiment, it's not good to run the microwave without something to absorb all the energy bouncing around--the metal walls and mesh on the door look absolutely shiny to the microwaves). Also place a little dab of water on the metal floor of the oven, and a comparably-sized dab up on the food tray, well away from the floor. See which one evaporates first. If the ant gets greedy and goes for the food, my prediction is that it'll get toasted.
Nah, ants are just well hard innit!
