When compared to many other sensitive methods for material detection, such as inductively coupled mass spectroscopy and thermal ionization mass spectroscopy, laser-induced breakdown spectroscopy (LIBS) typically exhibits a lower signal-to-noise ratio (SNR), resulting in higher detection limits. Increasing the SNR of LIBS would improve the ability to characterize the sample composition with increased accuracy and speed and reduce the amount of material needed to perform analysis. We have been investigating the effect of simple ultrashort laser pulse shaping on the SNR of LIBS. Our goal is to control the dynamics of the ionization and recombination processes in the laser-produced plasma to favorably affect the SNR associated with the line emission from the plasma. Pulse shaping is performed using an acousto-optic programmable dispersive filter. An adaptive learning algorithm is being developed to automate the pulse shape optimization process for maximization of LIBS SNR in nuclear security-relevant material characterization scenarios. We report a 27 % increase of the SNR for non-gated LIBS measurements of uranium by utilizing simple pulse shaping limited exclusively to excess quadratic spectral phase of the laser pulse.