Does legalizing marijuana result in decreased narcotic use as legalization proponents and some researchers claim? Or, conversely, does legalizing marijuana provide a gateway for experimentation and future “hard” drug use as critics of legalization and proponents of the gateway hypothesis attest? Now that several states in the U.S. have legalized marijuana for recreational use, it is possible to begin assessing the validity of these competing claims. Relying on a novel strategy for tracking trends in illegal drug use, we use Internet search queries, specifically Google Trends, to examine patterns of drug searches in four recreational marijuana states: Alaska, Colorado, Oregon and Washington. We find that search trend patterns for drugs are consistent following legalization, with increased mean searches for marijuana, cocaine and heroin, and decreased mean searches for methamphetamines, Oxycodone, and bath salts. While this finding seems to suggest some validity to the gateway hypothesis, we argue the opposite, for these general trends are also found at the national level. As such, the trends found in these recreational marijuana states generally do not differ from the nation as a whole, implying no significant gateway effect. We conclude that using big data to assess drug-using trends can inform the ongoing marijuana legalization debate.