The Dravido-Koreanic hypothesis is probably a bad example as it’s considered to be complete quackery by pretty much everyone who studies language. It was first suggested in 1905 and was based off the Turanian language hypothesis, which was found to be completely incorrect long ago.