Published paleointensity estimates derived from lavas extruded in known fields are highly variable and rarely recover the expected field strength within an accuracy of better than 10%. Inconsistent results on modern volcanic rocks lend even greater uncertainty to intensity experiments performed on lava flows emplaced during periods of unknown geomagnetic field strength. The majority of published paleointensity data are collected from the slowly cooled, massive centers of lava flows, where the magnetic grains are more likely to be multi-domain and produce non-ideal experimental results. Glassy volcanic material (found on subaerial lava flow tops and in sub-aqueous and subglacial environments), however is rapidly cooled, and therefore most likely of all volcanic materials to behave as single-domain particles demanded by Néel theory. We present a new paleointensity study of historic and modern Hawaiian lavas and test the viability of subaerially emplaced glassy basaltic material as an accurate recorder of magnetic field intensity. Six of eight lava flows sampled on the Big Island of Hawaii (1843, 1859, 1935, 1950, 1960, 1990 C.E.) produce well behaved Arai plots and recover an average intensity to within 2.7. μT of the expected field strength or better than 8% accuracy. We apply very strict selection criteria, including a minimum of three specimens per site, to prevent extraneous field estimates from affecting the final results. Individual volcanic glass results from the 1960 C.E. lava flow have a much lower variance than published data from the same volcanic unit. Glassy materials should therefore be collected wherever possible as they allow recovery of geomagnetic field strength with unprecedented accuracy.