The ability to generate regionally defined neuronal populations from human pluripotent stem cells (hPSCs) provides an important new experimental resource for the investigation of human neuronal physiology. We have developed a protocol in which H9 hPSCs are neuralised in suspension using chemically defined medium. Such an approach is based upon the default model of neurogenesis that minimizes extrinsic and intrinsic signals that lead to alternative cell fates. Analysis of the resulting neural stem cells (NSCs) reveals a forebrain identity as assessed by FOXG1 and OTX2 expression. Subsequently NSCs are plated as a monolayer to generate PAX6-positive precursors that terminally differentiate into an enriched VGLUT1 positive population of cortical neurones that express the cortical layer specific markers CTIP2, SAT2B or Reelin. Importantly, glia and inhibitory interneurones are only nominally present (<10% of total population). In the present study we have used this population of human cortical excitatory neurones (hCENs) to assess the developmental expression and biophysical properties of 2-amino-3-hydroxy-5-methyl-isoxazol-4-yl propanoic acid receptors (AMPARs). The responsiveness of the cells to AMPA (50 μM) was assessed weekly for 5 weeks following differentiation and significant increases in current densities from week 1 (0.12 ± 0.08 pA/pF, n = 22) to week 5 (1.14 ± 0.2 pA/pF, n = 35) were observed. As would be anticipated AMPAR-mediated currents were blocked by 6-cyano-7-nitroquinoxaline-2,3-dione (CNQX; 5 μM). Non-stationary noise analysis indicated that the mean weighted single-channel conductance of AMPARs expressed by hCENs decreased from 11.1 ± 1.2 pS (week 2; n = 8) to 4.5 ± 0.4 pS (week 5; n = 12). In addition AMPAR-mediated currents at week 5 were insensitive to intracellular block by the polyamine, spermine (100 μM; n = 5). Similar experiments with neurones derived from the 31D1 induced pluripotent stem cell line also indicated the presence of low-conductance AMPARs at week 5 (4.1 ± 1.2 pS; n = 3). Our data are consistent with the notion that at week 5 hCENs express AMPARs that contain the edited form of the GluA2 subunit. Furthermore, the developmental switch from an AMPAR population possessing higher unitary conductances to one expressing lower unitary conductances is tetrodotoxin insensitive (i.e. action potential independent). In addition this switch is not blocked when neurones are cultured in presence of antagonists of ionotropic glutamate receptors.