Unique challenges are faced in modelling faults in intraplate regions for seismic hazard purposes. Low fault slip rates compared to landscape modification rates lead to often poor discoverability of fault sources, and favours incomplete characterisation of rupture behaviours. Irrespective, regional and local test cases have demonstrated that fault sources assigned activity rates consistent with paleoseismic observations have the potential to significantly impact probabilistic seismic hazard assessments in Australia. To reflect this, the 2018 Australian NSHA will for the first time incorporate a fault source model. The model includes over 300 onshore faults, and a handful of offshore faults, which are modelled as simplified planes and assigned a general dip and dip direction. Dips are obtained from seismic-reflection profiles, where available, or inferred by taking into account surface geology and geomorphology, or other fault geometries within similar neotectonic settings. The base of faulting is generally taken as the regional maximum depth of distributed seismicity. Slip rates are calculated from displaced strata of known age, estimated from surface expression, or are extrapolated from other faults within similar neotectonic settings. We construct logic trees to capture epistemic uncertainty in fault source parameters, including magnitude frequency distribution, and the potential for random, periodic or episodic recurrence behaviour. This presentation introduces the new fault source database, the fault source logic tree as it currently exists, and discusses uncertainty in and sensitivity to various elements of the proposed fault source input model.