They may come to Australia next: http://www.ecdonline.com.au/content...-the-risk-of-electrical-fires-1188033210

High-frequency noise is mentioned as a way of detection, but how are they going to reliably ignore noise from things like switching power supplies and brush motors? (Sure, such items have filters; but they don't completely block RF, and the filter capacitors can deteriorate and drop in capacitance.)

And if they're so confident in the efficacy of AFCIs, why can't they come up with a standardised test?

I'd just want to see independent proof of their performance in realistic scenarios, to be convinced. (I can't say I've succeeded in maintaining an arc between a normal 240V plug and socket under load, so make of that what you will...)