One of the main operational aims of the MAST experiment (Darke et al 1994 Proc. 18th Symp. on Fusion Technology (Karlsruhe, Germany, 1994) p 799) and the proposed MAST upgrade is to investigate possible mechanisms to control the q-profile and drive off-axis current. Experiments were carried out to determine the extent to which the q-profile may be modified using two different approaches, transient and steady-state. Transient effects during the plasma current ramp-up phase were investigated with the aim of developing a start-up regime that can later be used as a target plasma for non-inductive current drive or to access advanced modes of operation such as the hybrid or improved H-mode. The most significant effect in this case was observed when early neutral beam injection (NBI) was applied to the fast current ramp-rate start-up plasmas causing reversed magnetic shear and the plasma current to 'pile-up' off-axis.In steady-state experiments, in which off-axis NBI was studied, results indicate that broadening the fast ion deposition profile by off-axis neutral beam (NB) injection helps to avoid harmful plasma instabilities and significantly extends the operational window of MAST. Long pulse (>0.65 s) H-mode plasmas were achieved with plasma duration limited only by present machine and NBI engineering limits. In order to match the experimentally observed neutron rate and stored energy a low level of anomalous fast ion diffuse ion (Db ∼ 0.5 m2 s−1) is required. The introduction of the fast ion diffusion broadens the neutral beam current drive profile and degrades the relative contribution of NB driven current from ∼40% to ∼30%. To obtain direct measurements of the current profile, a multi-chord motional stark effect diagnostic has been commissioned on MAST and is currently delivering first results in order to confirm the off-axis location of the NB driven current.