This dataset provides first results of the ongoing search for cryptotephras in the International Continental Scientific Drilling Program (ICDP) sediment core 5017-1 retrieved from the deep northern Dead Sea (DS) basin in 2010/11. 56 samples from sediments and rock salts were collected between 89.30 and 94.09 m sediment depth below lake floor from cores 5017-1-A-43 and 5017-1-A-44, focusing on the Lateglacial time period (~15-11.4 cal. ka BP), when Lake Lisan – the precursor of the DS – shrank from its glacial high-stand to the Holocene low levels. Sampling was performed in contiguous 5 cm steps with sample volumes of 5 cm³ and excluding mass-transported deposits thicker than 5 cm. The cryptotephra glass-shard extraction protocol followed established separation procedures (Blockley et al. 2005, doi:10.1016/j.quascirev.2004.12.008), and has been further adapted to the extreme salinity and sediment recycling of the DS. Glass shards were picked using a 100 μm-diameter gas-chromatography syringe attached to a micromanipulator (Lane et al. 2014, doi:10.1016/j.jas.2013.10.033), embedded in Araldite 2020 epoxy resin and ground and polished for electron probe microanalyses (EPMA). Major-element composition of individual glass shards was measured using a JEOL JXA-8230 electron microprobe at GFZ Potsdam, Germany (15 kV, 5-10 nA, 5-10 µm beam size). Instrumental calibration used natural mineral standards and analytical runs were monitored using glass standards (see second dataset). From six glass samples at least 10 glass shards per sample were measured by EPMA and for three samples replicate measurements on different glass shards were performed. After removal of glass geochemical measurements with totals <90%, 102 glass shard measurements remain in total. In general, cryptotephra is abundant in the Dead Sea record (up to ~100 shards per cm³), but often glasses are physically and/or chemically altered. The glass shard measurements reveal a heterogeneous geochemical composition, with mainly rhyolitic and some trachytic glasses potentially sourced from Italian, Aegean and Anatolian volcanoes. These first results demonstrate the great potential of crypto-tephrochronology in the DS record for improving its chronology and connecting the Levantine region to the Mediterranean tephra framework.