Donna Lee Elm
Google’s location history data are prized by law enforcement because they can be highly precise. Instead of the indeterminant and broad scope of cell tower location data, Google’s data can potentially identify a device’s location to within a yard or two. Moreover, Google tracked devices’ location quite often (sometimes as frequently as every two minutes), so the data can show movement and direction again with far greater precision than cell tower data. Google had been receiving warrants for its location history information on specified individuals’ devices, but in 2016, law enforcement recognized a new way to use those data.
The FBI for the first time sought what has come to be known as Geofence or Reverse-Location search warrants. Instead of identifying an individual whose data is sought, these warrants identified a place and a time where a crime occurred and sought information about who was there then. The government required Google to turn over information revealing all devices that were present in a certain geographic area at a particular period in time. Google imposed some limits on this process, requiring law enforcement to reasonably narrow its geographic area (the geofence) as well as the time frame to be searched. Then it established a three-step process to winnow down the number of “hits” that were produced before giving out identifying information; it sought to limit the invasion of its users’ privacy. Courts authorizing these warrants happily embraced that practice. But as discussed in my earlier article, see Donna Lee Elm, Geofence Warrants: Challenging Digital Dragnets, 35 Crim. Just. 2 (Summer 2020), these warrants have been criticized as being inherently overbroad, ignore the “particularity” requirement of the Fourth Amendment’s warrants clause, and, in the majority of instances, fail to actually identify the suspect.