In 2019, police investigating a credit-union robbery outside Richmond, Virginia, asked Google for the names behind every phone that had pinged inside a 150-meter radius around the scene across a one-hour window. Within three steps, an anonymous device sweep, a widened time window on each side of the robbery, and a final subscriber lookup, Google produced enough data to narrow the suspect pool to three people, then one: Okello Chatrie. He was convicted. Last week, the Supreme Court ruled 6-3 in Chatrie v. United States that this kind of request, a "geofence warrant," is now a "search" under the Fourth Amendment. The decision is narrower than the headlines suggest: the practice isn't banned. Police just need probable cause now.
Geofence warrants are the police tool that compels a tech provider to hand over information about every electronic device that was inside a defined geographic area during a defined window of time. Google maintains a large repository of anonymized location history known internally as Sensor Vault, and its "reverse-location" practice lets investigators start with a place and time and work backward to identify people. In the Chatrie case, the warrant swept in every phone that pinged within 150 meters of the credit union from 30 minutes before to 30 minutes after the robbery, according to Gizmodo's coverage. Google produced anonymized data first, then expanded the time window to two hours on each side and pulled a narrower device set, and finally handed over names and phone numbers for the people inside it. That trail led police to Chatrie, and a search of his home turned up the gun and roughly $100,000 in cash connected to the robbery, Gizmodo reports.
The majority's reasoning extends the logic of Carpenter v. United States, the Court's 2018 decision that historical cell-site location data is itself a search because it exposes where a person has actually been. Chatrie pushes that principle one step further: even when police don't know in advance whose phones they are pulling, asking a provider to map who was in a place is a search, because cellphone location data reveals the physical movements of the people carrying them. The full holding is in the slip opinion for Chatrie v. United States, No. 25-112, and Oyez's case page tracks the docket.
But the Court stopped short of requiring a warrant, the higher bar that would have demanded a judge's sign-off based on probable cause. Probable cause, the standard the Court applied here, lets police proceed if a reasonable person would conclude a crime had been committed and the suspect was involved. The Court has not yet said what that showing must look like for a bulk area query, where the device pool is large and most of the people swept in aren't suspects at all. SCOTUSblog's coverage flags the open question, and The Hill notes the practical effect: thousands of geofence requests a year will now face a higher legal threshold.
That gap is exactly where civil-liberties advocates say the fight isn't over. The Electronic Frontier Foundation's Andrew Crocker told reporters the ruling "didn't go far enough," because the bulk-query practice, in which police cast a wide net and narrow later, remains in an unsettled legal zone, according to Gizmodo. The Court took the cleaner doctrinal step, framing geofence warrants as searches, without answering the harder one: what probable cause actually requires when the device pool includes every phone that happened to be near a crime scene.
The practical floor just moved: from a low-bar administrative request to Google, to a probable-cause showing before a judge. What that showing must look like, and whether future cases will push the Court toward a full warrant requirement, is the question the next round of geofence litigation will have to answer.