Abstract Ever since Paul K. Benedict (1942, 1975) had first set out the case for Austro-Tai in detail, the latest published works to have summed up arguments for a genetic connection between Austronesian and Tai-Kadai (Kra-Dai) are Ostapirat (2005) and Sagart (2005), with Norquest (2013) providing more Austro-Tai lexical and phonological correspondences. Since 2005, there has been a huge amount of new Tai-Kadai data published in China, especially on the conservative and little-known Kra branch. Other than the Chinese new data, this study will take into account the various new and significantly revised reconstructions of mainland Southeast Asian language families and branches, namely Proto-Hlai by Norquest (2007), Proto-Tai by Pittayaporn (2009), Proto-Hmong-Mien by Ratliff (2010), and Proto-Mon-Khmer by Shorto (2006). Using this new evidence, I will demonstrate that some Tai-Kadai lexical forms are found only in certain divergent Formosan languages such as Rukai and Puyuma, but not in other Austronesian languages. The sporadic occurrences of some forms found only in Tai-Kadai and Malayo-Polynesian, such as *manuk 'bird', will also be discussed. The primary purpose of this data-intensive paper will be to examine the external etymological origins and links of Tai-Kadai lexical forms at various levels, especially the primary branch level. An extensive database will be used, and lexical isoglosses will also be identified and analyzed across different phyla and branches. More than simply another Austronesian sister or branch that has undergone significant Sinification, Tai-Kadai will be shown to have many basic vocabulary items shared with other mainland Southeast Asian phyla, namely Hmong-Mien, Austroasiatic, and Tibeto-Burman. While not necessarily evidence for common genetic inheritance, these forms indicate a complex linguistic scenario in prehistoric southern China, with extensive borrowing among various linguistic phyla within a linguistic area characterized by shared typological features such as sesquisyllabic nominal morphology. Thus, rather than simply cherry-picking for likely Austro-Tai cognates, this study will also look at the Tai-Kadai vocabulary that is not shared with Austronesian, and what the other types of evidence can elucidate about Southeast Asian prehistory.