Hey You, Get Out of My Cloud Data!: A Property-Based Approach to Reverse Search Warrants

Currently, there exists a “Grand Canyon” of judicial disagreement between the Fourth and Fifth Circuits on whether reverse search warrants (1) are searches under the Fourth Amendment and (2) are constitutional. Reverse search warrants are digital dragnets in which Google or other internet companies query massive amounts of data for the government to find someone suspected of a crime. For example, to determine who robbed a credit union in Midlothian, Virginia, the government asked Google for a “geofence” warrant in which Google scanned through the location history information of 592 million people stored on Google’s servers. Reverse search warrants also include keyword warrants, in which data companies like Google examine trillions of user searches to find individuals the government suspects of criminal activity.

This issue, on which state supreme courts are also divided, is now before the U.S. Supreme Court and awaiting decision as a question of Fourth Amendment jurisprudence. The Fourth Amendment protects from unreasonable searches and seizures. Since the advent of the modern “reasonable expectation of privacy” framework for searches and seizures in the late 1960s, courts have struggled with how to apply this test, resulting in a patchwork of constitutional protections. The disagreement and ensuing lack of clarity between the Fourth Circuit and the Fifth Circuit is emblematic of the varying results of the “reasonable expectation of privacy” framework. In the Fourth Circuit, a district court found that reverse warrants lack probable cause but that the good faith exclusionary rule applied. A panel of Fourth Circuit judges then held reverse warrants not to be a search under the Fourth Amendment, and the en banc Fourth Circuit later affirmed the district court in a fractured, one-sentence decision with eight different concurring opinions and no clear majority. The en banc court was divided as to whether a search occurred. On the other hand, the Fifth Circuit found reverse warrants to be a search and to be a constitutionally banned “general warrant.”

This Note proposes a different way of bringing clarity to this judicial disagreement. Instead of falling back on the “reasonable expectation of privacy” approach, courts should follow a property-based approach when it comes to questions of user data stored on the servers of internet companies. The property conception would provide a different path to resolve the disagreement by concluding that users have property-like protections in the data that reverse search warrants target and that this data qualifies as a “paper” under the Fourth Amendment, and thus the government must get a warrant to obtain this data.

This Note makes two main contributions. First, it surveys and analyzes the state of reverse search warrants in federal and state courts. Second, this Note proposes applying a property-based concept of privacy to the constitutionality of reverse search warrants, a concept that courts and academic literature have explored less than the “reasonable expectation of privacy” approach.

Part I of this Note provides an overview of what reverse search warrants are and how geofence and keyword warrants work on a technological level. Part II outlines a history of the Fourth Amendment and its interactions with emerging technology. Part III analyzes the procedural history and ensuing divide between the Fourth and Fifth Circuits and discusses the Fourth Amendment approaches state courts have applied to reverse search warrants. Part IV argues for a property-based theory of a search under the Fourth Amendment, but crucially not at the expense of the “reasonable expectation of privacy” framework. Following the lead of the Colorado Supreme Court in People v. Seymour, this Note proposes that users have a possessory interest in their data, anchored in the right to exclude.

Table of Contents Show

Introduction

Filmmakers love safe deposit boxes as a storytelling tool.[1] Audiences intuitively understand that these lockboxes are a symbol of near-impregnable security.[2] A safe deposit box, whether filled with jewels,[3] passports,[4] or cash,[5] represents the pinnacle of security in an analog world. When a character puts something into a safe deposit box (or another character breaks into the box), viewers know that something critical is happening.

In an increasingly digital world, safe deposit boxes may seem charmingly anachronistic outside of Hollywood.[6] However, the value of safe deposit boxes remains a useful way to think about the “new gold” (or oil)[7] of the twenty-first century: data, or more specifically, user data, living on the servers of internet companies.[8] In the context of the Fourth Amendment and reverse search warrants, courts should step in and recognize the power of user data. Courts should also recognize, in the Fourth Amendment context, that users own their data held by internet companies, much like depositors own their valuables in a safe deposit box in a bank. The law struggles to keep pace with technology, but applying the age-old concepts of property law to reverse search warrants will add additional protection to the Fourth Amendment for our digital future.

As Justice Sotomayor noted in her concurrence in United States v. Jones, in the “digital age . . . people reveal a great deal of information about themselves to third parties [internet companies] in the course of carrying out mundane tasks.”[9] Indeed, users give their data away in exchange for “free” services offered by internet companies.[10] In doing so, users may not know that they are sharing information like their current location or their searches on Google with countless internet companies. Users might also not know that internet companies might share their user data with law enforcement in response to novel investigative tools like reverse search warrants.[11]

While users may not grasp the power of “mundane” data, law enforcement recognizes the value of using new technologies and data for solving crimes.[12] For example, through tools like geofence warrants, law enforcement can acquire the location data of all Google users in a geographically bounded (“geo-fenced”) area.[13] Geofence warrants, a subset of reverse search warrants, subvert the probable cause and particularity requirements of the Fourth Amendment by compelling internet companies to examine all user location data to find unknown individuals suspected of crimes.[14] For example, to find someone accused of robbing a credit union in Midlothian, Virginia, the police obtained a geofence warrant that compelled Google to turn over location history data for everyone in the vicinity of the credit union at the time of the robbery. In this situation, Google searched 592 million[15] user account records for the location history of one person accused of robbing the credit union.[16] Law enforcement then used this location history to find, indict, and convict someone who allegedly robbed the credit union.[17]

Whether reverse search warrants are searches under the Fourth Amendment and comply with the probable cause and particularity warrant requirements is an open question and subject to debate among scholars, practitioners, and judges.[18] This question is also the focus of a “Grand Canyon” of disagreement between the Fourth and Fifth Circuits.[19] In 2024, a three-judge panel of the Fourth Circuit held in United States v. Chatrie that a search pursuant to a geofence warrant is not a search under the Fourth Amendment.[20] The same year, the Fifth Circuit, in United States v. Smith, reached the opposite conclusion and held that a search pursuant to a geofence warrant is a Fourth Amendment search and that the Constitution categorically bars geofence warrants.[21] The Fourth Circuit reheard Chatrie en banc in January 2025.[22] The full Fourth Circuit further muddied the waters by affirming the district court’s decision (and vacating the panel’s opinion) in April 2025 in a single sentence and issuing eight concurrences and one dissent, with no clear ruling on whether a search had occurred.[23] The U.S. Supreme Court granted certiorari on the question of geofence constitutionality in January 2026.[24] State courts are also divided on the issue.[25]

Much of the existing literature[26] and litigation[27] focuses on whether a search based on a reverse warrant violates a user’s reasonable expectation of privacy (REOP) in the records, under the framework established in Katz v. United States, and elaborated upon in the digital data context in Carpenter v. United States.[28] The scholarship examining an alternative property approach to reverse search warrants is underdeveloped by comparison.[29] While advocates have raised property-based claims[30] grounded in Justice Gorsuch’s dissenting opinion in Carpenter v. United States,[31] lower courts have rarely engaged with property-based arguments.[32]

To fill this gap, this Note makes two main contributions. First, as the issue of reverse search warrants evolves quickly and is likely to make its way to the U.S. Supreme Court,[33] this Note surveys and analyzes the state of reverse search warrants in federal and state courts across the country. Second, this Note considers and proposes applying a property-based concept of privacy to the constitutionality of reverse warrants.

This Note focuses its analysis on applying the protections of property to data targeted by reverse search warrants.[34] Property-like protection is especially powerful in the context of reverse search warrants, where user data held by large internet companies lends itself to the property concept of bailment. This Note does not argue for a wholesale importation of property law to Fourth Amendment issues or for a strict originalist or textualist reading of the Fourth Amendment. Instead, through a pragmatic lens, this Note expands on the approach that the Colorado Supreme Court announced in People v. Seymour.

This Note examines several different potential sources of property-like protections for user data, including positive law,[35] general law,[36] and Terms of Service (ToS) agreements.[37] While a large divide exists among scholars about the utility of using property law for Fourth Amendment questions,[38] the mechanics of reverse warrants present an instance in which property-like protections expand Fourth Amendment protections, especially in light of the splits using the Katz approach. To indulge once again in the privacy lawyer’s favorite analogy: This Note argues that the government peering into a safe deposit box is always a search under the Fourth Amendment because the depositor, and not the bank, owns the items in the box. This Note proposes that a property-based conception and the application of property-like protections, as a supplement to the REOP approach, would resolve the disagreement on reverse warrants between the Fourth and Fifth Circuits and make plain the unconstitutionality of such dragnet searches. Combining several existing property frameworks may offer a new way to defend Fourth Amendment rights in the face of reverse warrants. The combined framework that this Note adopts asks three questions: First, are the data targeted by a reverse search warrant “papers” under the Fourth Amendment? Second, do the protections of property apply to the data (or papers)? And third, did a search or seizure of the data (or papers) occur under a property-based conception of the Fourth Amendment?

In Part I, this Note provides an overview of what reverse search warrants are and how geofence and keyword warrants work on a technological level. Part II outlines a history of the Fourth Amendment and its interactions with emerging technology. This Section pays special attention to the development of the REOP framework from Katz, the third-party doctrine from United States v. Miller and Smith v. Maryland, and the re-emergence of a property-based conception of privacy in United States v. Jones, Florida v. Jardines, and Justice Gorsuch’s dissent in Carpenter v. United States. Part III analyzes the divide between the Fourth[39] and Fifth[40] Circuits and discusses the approaches state courts[41] have taken to reverse search warrants. Part IV argues for a property-based theory of search under the Fourth Amendment in the context of reverse warrants, but not at the expense of the Katz framework. Following the lead of the Colorado Supreme Court in Seymour, this Note argues that users have a possessory interest in their data.[42] This Note also considers the strengths and shortcomings of various property-based approaches to the Fourth Amendment.[43]

I. Reverse Search Warrants: Geofence and Keyword Warrants

A.      How is a Reverse Search Warrant Different from a Regular Search Warrant?

If the government’s action is a search under either the REOP[44] or property frameworks,[45] the Fourth Amendment warrant requirement demands that the government acquire a warrant to execute the search, unless an exception applies.[46] Courts understand the warrant requirement as several hurdles including (1) probable cause;[47] (2) a neutral and detached magistrate;[48] and (3) particularity.[49]

In a regular search warrant, the prosecution states the basis for its probable cause and gives a description of the items or people to be searched in its presentation to a magistrate or judge, who then approves or denies the warrant.[50]

In a reverse search warrant, the government points at a massive dataset and states that probable cause and particularity exist because of the size of the data, without specifically identifying either.[51] Reverse warrants are a recent development, with federal agents first using them in 2016.[52] Reverse warrants fall into two main categories: geofence and keyword warrants.[53]

Reverse Search Warrants[54]   Warrants for searches through massive digital databases often without a targeted individual to uncover potential suspects.[55]   Follow a multistep process (huge data request from law enforcement; return “anonymized” data; law enforcement reviews data; requests additional or de-anonymized data).[56]

Geofence Warrants   Subset of reverse search warrants.[57]   Target: location history data.   Potentially search through hundreds of millions of users’ data.[58]   Google has received the bulk of these warrants, but any internet company that retains user location history information could receive warrant.[59]   Thousands of warrants served on Google from 2016–2022.[60] Google claims to no longer store data in a format accessible to geofence warrants.[61]

Keyword Search Warrants   Subset of reverse search warrants.[62]   Target: user search history data (for example, what users type into Google) but could be for any company that retains search history info.[63]   Potentially search through billions of users’ data.[64]   Unknown how many keyword warrants exist.

1.       Geofence Warrants

In a geofence warrant, government investigators go to a data provider, typically Google,[65] with a search warrant for the location history of every Google account user in a geographically bound (or “geofenced”) area.[66] Google received thousands of geofence warrants from 2016–2020 (the most recent year data is available) and, at times, geofence warrants made up as much as 25 percent of all U.S. warrant requests to Google.[67] Most of the requests came from local or state law enforcement, with California requestors making up the largest percentage at approximately 17 percent of requests.[68]

When a user “opts”[69] into location history data, Google constantly collects, tracks, and stores this location data, both for the user’s benefit (allowing the user to review where they have been) and for advertising purposes.[70] In a thorough summary, the district court in Chatrie recounted that location history data comes from multiple sources including “[GPS] information, Bluetooth beacons, cell phone location information from nearby cellular towers, [IP] address information, and the signal strength of nearby Wi-Fi networks.”[71] Furthermore, a Google employee testified in an evidentiary hearing in United States v. Chatrie in 2022 that Google logs this information once every two minutes for each device.[72] This data goes into a database Google calls the “Sensorvault,” where each data point is associated with a unique user account.[73] The Sensorvault is massive—the court in Smith wrote that it contains information on 592 million people.[74] Notably, Google maintains that data stored in the Sensorvault remains under the control of the user.[75] In its amicus brief in support of neither party in Chatrie, Google wrote: “[I]t is the user who controls the [location history] information. The user can review, edit, or delete her Timeline and [location history] information from Google’s servers at will.”[76]

In consultation with the Computer Crimes and Intellectual Property Section of the Department of Justice, Google created a three-step process for responding to geofence warrants.[77] In step one, law enforcement gets a warrant requesting a “de-identified” list of all Google users within a geofenced area in a specific period.[78] Google maintains that the data produced in step one is “de-identified,” as it relates only to the location of the cell phone and does not contain names, emails, or other personally identifying information.[79] It appears to be an open question whether the data is actually “de-identified,” as a defense expert in United States v. Chatrie demonstrated that he was able to identify users from step-one data with a few simple public records searches.[80] As part of this step, Google must query the entire Sensorvault to narrow down a list of users within the parameters of the warrant.[81] In step two, law enforcement reviews the list of de-identified users to see if any are relevant to the investigation.[82] At step three, law enforcement compels Google to provide identifying information on the accounts of interest from step two.[83]

Google claims it challenges warrants it finds to be overly broad.[84] A Bloomberg Businessweek analysis of Google search warrant records from 2020–2023 found that police departments requested geofence warrants to solve routine crimes, like theft.[85] Also of note is that the people responsible for processing Google’s responses to both geofence and keyword warrants are members of the Legal Investigations Support team.[86] Legal Investigations Support members are often young professionals who “view the role as a way to test the waters before applying to law school.”[87] As a result, “[t]here’s a whole field of common law surveillance being written every day by twentysomething-year-olds.”[88] In this way, access to intimate data has been delegated to junior employees at a private company. This “common law” of surveillance extends to Google’s responses to keyword search warrants.

In December 2023, Google updated its location history policy and stated that location history data would be stored on the user’s device, set to delete after three months, and encrypted if the user chose to upload it to the cloud.[89] However, Google’s actions do not moot the constitutional concerns raised by the thousands of geofence warrants it received from 2016–2020 (and the unknown quantity presumably received from 2020–2023). Furthermore, Google never acknowledged it would no longer respond to geofence warrants. Rather, advocates infer that it would  be harder, if not impossible, for Google to respond to geofence warrants based on the changes Google made to its location history data storage.[90] As recently as the summer of 2025, local law enforcement continues to ask FBI counterparts for geofence warrants.[91] Additionally, the government’s briefing to the U.S. Supreme Court in Chatrie represents that Google has deleted data in the Sensorvault, except for information subject to litigation holds, demonstrating the continued viability of reverse warrant litigation.[92]

Other companies like X (formerly Twitter) have raised concerns that they may also be recipients of reverse search warrants because they “collect[], process[], and store[] multiple classes of sensitive user data,” including location data.[93] Thus, any entity that maintains user data could be subject to a reverse search warrant, indicating the far-reaching consequences of the disagreement between the Fourth and Fifth Circuits.

2.       Keyword Search Warrants

Every year, users enter more than five trillion searches into Google.[94] Google expects this number to increase with the addition of AI and visual searches.[95] As any savvy internet user understands, what one types into a search bar (or takes a picture of) does not disappear after the query is complete: Google retains records of each search its users enter.[96] A keyword search warrant is a type of reverse search warrant that compels Google to trawl through its vast database of searches to identify individual users who made a specific query.[97] Query examples range from mundane things like “weather”[98] or “youtube”[99] to highly personal information like “buy abortion pills.”[100] For example, in Latice Fisher’s case in Mississippi, law enforcement used her Google search results of “buy abortion pills, mifepristone online, misoprostol online,” and “buy Misoprostol abortion pill online” to charge her with second-degree murder after she experienced a stillbirth.[101]

Both law enforcement and Google understand the power of search results and keyword warrants.[102] In an effort to protect user privacy, Google has created a multistep process to respond to keyword search warrants in a similar fashion as they do for geofence warrants.[103] In step one, law enforcement presents Google with a keyword warrant that asks for a list of all searches for a certain term within a specific timeframe.[104] For example, in People v. Seymour the Denver Police Department asked Google for “a list of any users who had searched one of nine variations of ‘5312 N. Truckee St.,’” the address of a suspected arson, in the fifteen days prior to the arson.[105] Google then provided an “anonymized list of responsive searches” along with IP address information.[106] Law enforcement reviewed this anonymized list to determine if any of them were relevant to their investigation and obtained a second warrant compelling Google to identify the users who made the relevant searches.[107] In Seymour, Google turned over a list of sixty-one searches made by eight accounts.[108] Law enforcement identified five of the accounts as having Colorado-based IP addresses and compelled Google to turn over identifying information (like user account names and email addresses) for these five accounts.[109] The Denver Police Department then made an arrest for the arson based on the information that Google disclosed.[110]

II. Legal Background

At the heart of reverse search warrant issues are two main questions: First, is a query of a database pursuant to a reverse search warrant a search under the meaning of the Fourth Amendment; and second, if it is a search, can the query comply with the Fourth Amendment’s warrant requirements? To address these issues, this Section will provide a historical and legal framework of Fourth Amendment jurisprudence.

A. Fourth Amendment Overview

The Fourth Amendment to the United States Constitution protects “[t]he right of the people to be secure in their persons, houses, papers, and effects” from “unreasonable searches and seizures.”[111] The “warrant requirement” of the Fourth Amendment sets out the guidelines for a reasonable search or seizure.[112] The warrant must be supported by (1) probable cause,[113] (2) oath or affirmation, and (3) be sufficiently particular (or in clearer, nonlawyer language, be specific enough to easily identify what the warrant targets) in describing “the place to be searched, and the person or things to be seized.”[114]

At the time of the framing of the Constitution, English law had enshrined the right to be free from unreasonable searches and seizures.[115] The Framers took inspiration from English law and their own experiences when drafting the Fourth Amendment.[116] To combat smuggling in the colonies, the Crown issued general warrants called writs of assistance that empowered an agent to enter any house or other place to search for smuggled goods and to command anyone to assist the agent.[117] The amendment was drafted to prohibit these practices.[118]

The Fourth Amendment has been a right with a limited remedy since the founding, but especially so for the first 120 years of the existence of the United States.[119] It was not until 1914 in Weeks v. United States that the U.S. Supreme Court announced the exclusionary rule, which allowed federal courts to keep out evidence gathered in violation of the Fourth Amendment.[120] As part of the Bill of Rights, the Fourth Amendment only applied to the federal government for much of its existence.[121] The U.S. Supreme Court has gradually incorporated the protections of the Bill of Rights and applied them to the states through the due process clause of the Fourteenth Amendment.[122] In 1949, in Wolf v. Colorado, the U.S. Supreme Court found that the Fourth Amendment applied to the states.[123] In 1961, the Court further held in Mapp v. Ohio that the exclusionary rule applied to evidence gathered illegally in state prosecutions.[124]

The limited nature of the rights and remedies recognized by the Fourth Amendment continues today. The prohibition against unreasonable searches[125] and seizures,[126] the warrant requirement,[127] and the exclusionary rule[128] are riddled with exceptions. The good-faith exception, as announced by the U.S. Supreme Court in United States v. Leon,[129] is particularly relevant because the district court[130] (and concurrences in the en banc Fourth Circuit)[131] in Chatrie and the Fifth Circuit relied on it in their reverse warrant decisions to justify not dismissing the defendants’ case.[132] This exception holds that if officers rely in good faith on a search warrant that a court later finds to be defective, the evidence is not barred by the exclusionary rule.[133]

Also notable in Leon is that the U.S. Supreme Court affirmed that the sole purpose of the exclusionary rule is to deter police misconduct.[134] Earlier courts had believed that the exclusionary rule served to deter misconduct and to ensure the integrity of the judicial process.[135] The Court significantly weakened the protections of the Fourth Amendment by ruling that the exclusionary rule only serves a deterrence function.[136] Furthermore, the U.S. Supreme Court set a high standard in Leon for conduct covered by the exclusionary rule: Evidence may be excluded if the police mislead the judge with information they “knew was false or would have known was false except for [their] reckless disregard for the truth.”[137]

Information that the police gather through violating the Fourth Amendment but for which exclusion would serve no deterrence purpose is not excluded.[138] This post-Leon contrast is especially stark in the geofence case United States v. Smith, where the Fifth Circuit held that reverse warrants are “general warrants categorically prohibited by the Fourth Amendment.”[139] Smith had been convicted of robbing a postal carrier and law enforcement had only found him due to a reverse warrant, in this case, a geofence warrant.[140] Even though Smith was identified and convicted through an unconstitutional reverse search warrant, the court found that the good-faith exception applied because granting Smith any relief (e.g., shortening his prison term or granting him a new trial) would not deter future law enforcement violations.[141] In the court’s view, the law enforcement officers acted reasonably considering the unsettled state of the law of reverse search warrants.[142] There is a cruel irony in the Fifth Circuit using the raw material of Smith’s case to change the law to make reverse search warrants unconstitutional—a decision that will help all future defendants but which did not grant Smith any relief in his own case. Moreover, the Smith decision will make it harder for law enforcement to rely on the good-faith exception in future cases in the Fifth Circuit in reverse warrant cases.[143] Law enforcement is now on notice that these kinds of warrants are unconstitutional, which makes their continued use unlikely to be a “good faith” mistake.[144]

Despite the checkered history of the Fourth Amendment, it remains a useful tool to guard against government overreach. Liberal and conservative justices alike are among its champions.[145] Since the dawn of the contemporary framework in Katz v. United States in 1967, Fourth Amendment search analysis has proceeded in a linear fashion. A reviewing court will ask (1) was there a search under the meaning of the Fourth Amendment; (2) if there was a search, does the search require a warrant or does it fall into one of the warrant exceptions; and (3) if the search did require a warrant, did the warrant meet the requirements of probable cause, a neutral and detached magistrate, and particularity?

B. What is a Search?

The U.S. Supreme Court has answered this question under the Fourth Amendment along two different axes: the reasonable expectation of privacy test from Katz v. United States and its progeny, and the property-based approach from United States v. Jones.

1.       The Reasonable Expectation of Privacy from Katz v. United States

In Katz,the U.S. Supreme Court declared that the Fourth Amendment “protects people, not places.”[146] The Court also announced that “[w]hat a person knowingly exposes to the public, even in his own home or office, is not a subject of Fourth Amendment protection,” demonstrating a theoretical approach that de-emphasized the historical primacy of property rights to the Fourth Amendment.[147] The Court rejected the idea that the Fourth Amendment is limited only to the search and seizure of “tangible property.”[148] In his concurrence, Justice Harlan formulated what has been called the “reasonable expectation of privacy” test.[149] REOP is the main test courts now use to determine if the government’s actions constitute a search.[150]

The REOP test sets forth two prongs for determining whether the government’s action is a Fourth Amendment search. First, does the person have a subjective expectation of privacy in whatever it is the government seeks to access;[151] and second, is society prepared to recognize that expectation as objectively reasonable?[152] For example, the facts of Katz (which seem as anachronistic to a twenty-first century reader as a physical safe deposit box) demonstrate that people have a reasonable expectation of privacy in places where they can reasonably assume they are alone and not subject to eavesdropping.[153] Katz, a prolific college basketball sports bettor, made use of a public payphone near his apartment in Los Angeles to place bets with people in Boston and Miami.[154] The FBI conducted an elaborate sting operation and placed a listening device on the outside of the phone booth to record Katz’s conversations.[155] The FBI arrested Katz for violating 18 U.S.C. § 1084, transmission of wagering information over the telephone.[156]

When Katz’s case made its way to the U.S. Supreme Court, the Court held that Katz had an objectively reasonable expectation of privacy in a public phone booth.[157] The Court dismissed the government’s arguments that the phone booth was not a “constitutionally protected location” and relied instead on the fact that a person who enters a telephone booth “shuts the door behind him, and pays the toll that permits him to place a call is surely entitled to assume that the words he utters into the mouthpiece will not be broadcast to the world.”[158] The FBI’s installation of a listening device on the outside of the phone booth intruded on Katz’s expectation of privacy. The Court arrived at this conclusion on its own and noted that the parties had spilled a great deal of ink briefing what was and was not a “constitutionally protected area.”[159] Over the past fifty years, courts have refined REOP to focus on whether a person had a societally reasonable expectation of privacy, with little regard for their subjective expectations.[160]

2.       Katz and Emerging Technology: The Third-Party Doctrine

Katz dealt with a then-emerging technology issue: an electronic listening device to record Katz’s end of conversations in a public payphone.[161] Legal scholars, judges, and lawyers have all hailed the REOP approach as an expansion of Fourth Amendment jurisprudence and criticized it for producing a hard-to-apply framework.[162]

The tension between REOP and emerging technology led to the emergence of the third-party doctrine, through a series of cases in the 1970s. The third-party doctrine holds that a person has no REOP in information they voluntarily give to a third party.[163] For example, under the third-party doctrine, if a person gives information to a bank related to her deposits and withdrawals simply by doing business with the bank, she has no REOP in that information since she shared it with the bank.[164] The government can then obtain information about her deposits and withdrawals without a warrant.[165]

Since its emergence, the third-party doctrine has been at odds with the realities of modern life.[166] For example, in Smith v. Maryland, the U.S. Supreme Court found that a “pen register” (a device installed at the telephone company that recorded the numbers dialed by a user, but not the contents of the conversation) was not a search because the user voluntarily transmitted the dialed number to the phone company and “assumed the risk that the company would reveal to police the numbers he dialed.”[167] But did the user voluntarily transmit the dialed number to the phone company? As Justice Stewart noted in his dissent, if the user did not dial the number, they wouldn’t be able to complete the call, seemingly negating the action’s voluntariness.[168]

The U.S. Supreme Court recognized limits to the third-party doctrine in the 2012 case United States v. Jones[169]and the 2018 case Carpenter v. United States.[170] In Carpenter, the U.S. Supreme Court held that the third-party doctrine does not apply to cell-site location information (CSLI) (information users share with wireless carriers). The Court noted that “carrying [a cell phone] is indispensable to participation in modern society” and that users do not voluntarily share CSLI with wireless carriers—CSLI is transmitted as part of the operation of the phone.[171] The Court also cautioned against “mechanically applying the third party doctrine” without considering the nature of the information shared.[172] The Carpenter Court characterized its decision as “narrow” and claimed not to “disturb the application of Smith and Miller or call into question conventional surveillance techniques.”[173] Professor Tokson identifies the “Carpenter Test” as consisting of three factors: the deeply revealing nature of location data, the quantity of data collected, and the fact that users do not voluntarily transmit CSLI.[174]

Advocates have hailed Carpenter as a revolution for privacy rights.[175] However, as the Fourth Circuit noted in Chatrie, a reverse warrant case, the third-party doctrine remains alive and well today.[176] The U.S. Supreme Court has also acknowledged that some digital technologies are categorically different than pre-digital technologies and that attempting to draw analogies between the two may be fruitless due to the sheer quantity of records one can store digitally, for example on a cell phone.[177]

REOP may have expanded Fourth Amendment protections beyond physical property, but it has led to inconsistent results, especially with emerging technology issues. This is prominently on display in the divide between the Fourth Circuit in Chatrie and the Fifth Circuit in Smith. Though it was vacated by the Fourth Circuit’s en banc rehearing, the Fourth Circuit panel held that “the third-party doctrine squarely governs” the use of geofence warrants,[178] while the Fifth Circuit held that “per Carpenter, the third-party doctrine does not apply” to geofence warrants.[179] Using the protections of property may fill some of the gaps the U.S. Supreme Court has created in REOP.

3.       A Re-Emergence of the Property Framework

In United States v. Jones, the U.S. Supreme Court held that FBI agents conducted a search when they placed a GPS tracker on Jones’s Jeep and used it to track his movements over twenty-eight days, generating more than two thousand pages of data.[180] The FBI and Washington D.C. police suspected that Jones, who owned a nightclub in D.C., was dealing drugs.[181] As part of their investigation, which also included visual surveillance of the club and wiretapping Jones’s cell phone, agents put a GPS beacon on the underside of Jones’s Jeep when it was parked in a public lot.[182] Instead of relying on the REOP framework, the majority opinion, authored by Justice Scalia, looked to property rights to find a Fourth Amendment violation.[183]

The Court in Jones found that a car is an “effect” (personal property)[184] in the meaning of the Fourth Amendment and that the government’s physical occupation of Jones’s car to place the GPS monitor on it was a search.[185] Justice Scalia wrote that the Katz test did not displace the common-law trespassory test, which the framers of the Constitution would have recognized at the time of the founding.[186] Similarly, in announcing a return to the trespassory test, Justice Scalia stated, “[W]e do not make trespass the exclusive test,” seemingly preserving the validity of the Katz framework.[187] Justice Scalia explained that “our Fourth Amendment jurisprudence was closely tied to common-law trespass, at least until the latter half of the 20th century.”[188] Trespass is defined as “knowingly entering another owners’ property or land without permission, which encroaches on the owners’ privacy or property interests.”[189] In Jones, Justice Scalia held that “[t]he Government physically occupied private property for the purpose of obtaining information” by placing the GPS tracker on Jones’s car, and thus that there was “no doubt that such a physical intrusion would have been considered a ‘search’ within the meaning of the Fourth Amendment when it was adopted.”[190]

Justice Scalia, writing for the Court, expanded the property-based conception of a search in the 2013 Florida v. Jardines decision.[191] In Jardines, the Court announced that the property-based approach included physical intrusion—a likely broader understanding than trespass.[192] Police officers had approached the area surrounding Jardines’s home with a drug-sniffing dog that alerted them to the presence of marijuana plants.[193] The Court, again relying on the text of the Fourth Amendment, wrote that “when it comes to the Fourth Amendment, the home is first among equals.”[194] The Court held that by bringing the drug-sniffing dog close to the home, and by not knocking to gain a license, the officers had impermissibly intruded on Jardines’s property rights.[195] The Court further explained that “[o]ne virtue of the Fourth Amendment’s property-rights baseline is that it keeps easy cases easy,” meaning that all the Court had to do to determine if a search occurred was to ask if the officers intruded on Jardines’s property.[196]

Most recently, and most importantly for the reverse warrant debate, a property-based theory of the Fourth Amendment emerged in Justice Gorsuch’s dissent in Carpenter.[197] Justice Gorsuch advocated for treating CSLI as a “bailment,” a type of property law relationship in which the user (or bailor), assigns their data to the company (bailee), without giving up ownership rights in the transferred property.[198] The appeal of this approach is clear: If the user owns the data, under the bailment approach to the Fourth Amendment, a search occurs if the government accesses the data without the user’s permission. Under this approach, courts do not need to engage in the legally complex REOP analysis.[199] Justice Gorsuch claims not to dismiss the Katz approach but offers bailment as a way of simplifying thorny Fourth Amendment issues.[200]

III. How Courts Think About Reverse Search Warrants

As of writing in January 2026, whether geofence warrants or reverse warrants more broadly are a search and whether they are constitutional is an unsettled question. Among federal courts, panels on the Fourth and Fifth Circuits have reached opposite conclusions. The Fourth Circuit panel in Chatrie held that geofence warrants are not a search.[201] The Fifth Circuit in Smith ruled that geofence warrants are a search and that they cannot comply with the Constitution’s warrant requirements.[202] Both Chatrie and Smith filed petitions for certiorari in the U.S. Supreme Court.[203] Chatrie focused on whether the geofence was constitutional and highlighted the circuit split in his petition for certiorari.[204] The Court granted the petition as to whether geofence warrants are constitutional on January 16, 2026.[205] Smith challenged the Fifth Circuit’s application of the good-faith exception, arguing that there can be no good-faith reliance on a general warrant.[206] The U.S. Supreme Court denied Smith’s petition for certiorari on November 10, 2025.[207]

The federal district court for the District of Columbia ruled that geofences are a search but that they can comply with the Constitution in United States v. Rhine—a January 6th case.[208] Following the Fifth Circuit’s binding precedent, a magistrate in the Southern District of Mississippi held a warrant for a “tower dump” to be a reverse search like a geofence warrant, in which law enforcement obtains the records of all cell phone users from a specific cell tower, to be categorically barred by the Fourth Amendment.[209] In the Ninth Circuit, Judge Du in the District of Nevada adopted the Fifth Circuit’s approach to reverse warrants, writing that “the court finds that a tower dump is a search and the warrant law enforcement used to get it is a general warrant forbidden under the Fourth Amendment.”[210] However, since this was the first time that a court within the Ninth Circuit ruled on a reverse search warrant issue, Judge Du also held that the good-faith exception applied and denied the defendant’s motion to suppress.[211]

State supreme courts have also weighed in, applying both federal constitutional principles and state constitutional law. In People v. Seymour, the Colorado Supreme Court issued a multifaceted decision touching on both the REOP and property conceptions of a search, ultimately holding that the keyword search was constitutional.[212] Likewise, in Commonwealth v. Kurtz, the Pennsylvania Supreme Court held that a user has no REOP in their internet search history and thus cannot challenge a keyword search warrant.[213] In Jones v. State, the Georgia Supreme Court held a geofence to be a search and that it complied with the Fourth Amendment.[214] Deepening the split, in April 2025, the Texas Court of Criminal Appeals held in Wellsv. State that the use of a geofence warrant did not violate the Fourth Amendment.[215] In Texas, there is now a split between the state court system and the federal court system on the constitutionality of geofence warrants. Wells has appealed his case to the U.S. Supreme Court on both the constitutionality of the geofence warrant and the good-faith exception issues; his petition for certiorari is pending as of February 2026.[216]

A. Federal Courts

1.       United States v. Chatrie (Fourth Circuit) and United States v. Smith (Fifth Circuit)

On similar sets of facts, panels in the Fourth and Fifth Circuits came to different conclusions about (1) whether a geofence is a search, and (2) if a geofence warrant is constitutional. In both Chatrie (Fourth Circuit) and Smith (Fifth Circuit), law enforcement turned to geofence warrants to identify people who committed robberies.[217] In Chatrie, a person armed with a firearm and wearing a hi-vis vest[218] robbed a credit union in Midlothian, Virginia.[219] In Smith, a person armed with a firearm robbed a mail carrier at a post office in rural Mississippi.[220] The amount taken from the credit union in Chatrie was $195,000.[221] In Smith, the person who robbed the mail carrier made off with $60,706 from three mail sacks and the mail carrier’s post office keys.[222]

In both Chatrie and Smith, investigating officers reviewed security camera footage and noted that the people who committed the robberies appeared to be using a cell phone at some point in the course of the robbery.[223] In both instances, law enforcement seized on this detail and turned to geofence warrants after allegedly exhausting all other investigatory leads.[224] Law enforcement in Chatrie made their first geofence warrant request three weeks after the robbery.[225] In Smith, law enforcement requested a geofence warrant nine months after the robbery.[226] Google received the warrants, and after following its three-step process for responding to geofence warrants, produced location data that law enforcement used to identify Chatrie[227] and Smith.[228]

In Chatrie and Smith, the defendants were charged in federal district court and filed motions to suppress the evidence collected from the geofence warrants (the Google location data), arguing that law enforcement obtained the evidence in violation of the Fourth Amendment.[229] Here is where the legal trajectories of Chatrie and Smith diverge.

In Chatrie, the district court held that a geofence warrant is a search under the Fourth Amendment and that the warrant in this case violated the Fourth Amendment.[230] The district court held that the warrant was not supported by probable cause or particularity and was therefore invalid.[231] However, it also held that the good-faith exception to the warrant requirement applied.[232] It examined the warrant only under the REOP framework, even though the defendant submitted both property-based and REOP arguments in his brief.[233]

In contrast, the district court in Smith declined to rule on whether a geofence warrant is a search.[234] The district court held that the warrant in the case had sufficient probable cause and it was sufficiently particular for step one of Google’s geofence warrant response process.[235] The court found further that the warrant was not sufficient for steps two and three, and thus an additional warrant was required.[236] Returning to a point of convergence with the Chatrie court, the Smith court also found that the good-faith exception applied.[237]

In both Smith and Chatrie, the defendants appealed the district courts’ decisions. The Fourth and Fifth Circuits offered extraordinarily different opinions, initially creating a “Grand Canyon” of circuit splits.[238]

In Chatrie, a panel of the Fourth Circuit, in a 2-1 decision, held that geofence warrants are not searches under the Fourth Amendment.[239] Like the district court, the Fourth Circuit panel analyzed the issue only in terms of REOP.[240] The panel concluded that the information the government sought was limited in time, as it only obtained two hours’ worth of location history data.[241] Analogizing to Miller, the majority wrote: “A record of a person’s single, brief trip is no more revealing than his bank records or telephone call logs,” which seemingly underestimates the revealing nature of bank records and phone logs.[242] The Fourth Circuit also reasoned that users voluntarily share their location history with Google and that their consent is made clear through the on-screen prompts to enable location history.[243] As a result of the limited nature of the data and the voluntariness of the exposure, the Fourth Circuit concluded that the third-party doctrine applies to location history information, and thus users have no REOP in the data.[244] In the dissent, Judge Wynn highlighted the absurdity of the panel’s result: The government itself believed a geofence search to be a search under the Fourth Amendment because it sought a warrant in the first place.[245]

The defendant in Chatrie sought and obtained a rehearing before the full Fourth Circuit in early 2025.[246] The full decision came out on April 30, 2025, and it did nothing to bridge the divide between the Fourth and Fifth Circuits.[247] The full decision states: “PER CURIAM: The judgment of the district court is AFFIRMED.”[248] To complicate matters further, the Fourth Circuit judges wrote eight concurring opinions, and none of them received a majority of votes.[249] According to Professor Orin Kerr, the fractured decision means “there is no majority opinion, but instead just a crazy amount of uncertainty. What is the law now, after all this? I haven’t a clue.”[250]

On the other hand, in Smith, a unanimous Fifth Circuit panel held that geofence warrants are searches under the Fourth Amendment and that due to their very nature, they can never comply with the warrant requirement of the Fourth Amendment.[251] Judge Ho, in his concurrence, made plain the panel’s reasoning: “[O]ur panel decision today will inevitably hamper legitimate law enforcement interests. But hamstringing the government is the whole point of our Constitution.”[252] The Fifth Circuit also held that the good-faith exception applied and upheld Smith’s conviction.[253] Like the Fourth Circuit, the Fifth Circuit analyzed the warrant through an REOP lens but came to the opposite conclusion.[254] Here, the Fifth Circuit leaned heavily on Carpenter, reasoning that “[p]erhaps the most alarming aspect of geofences is the potential for permeating police surveillance,” and disagreeing with the Fourth Circuit on the factual basis of the pervasiveness and granularity of digital surveillance capabilities.[255] The Fifth Circuit found that the “opt in” process to sharing location history data with Google was convoluted to the point of not being voluntary, and thus the third-party doctrine did not apply.[256]

The Fifth Circuit went beyond its REOP analysis to find that geofence warrants are unconstitutional general warrants.[257] The court stated: “These geofence warrants fail at Step 1—they allow law enforcement to rummage through troves of location data from hundreds of millions of Google users without any description of the particular suspect or suspects to be found.”[258] The government attempted to have Smith reheard en banc, but the Fifth Circuit denied this in January 2025.[259]

B.      State Courts

Deepening this rift, state supreme courts have also weighed in on the reverse warrant issue and have applied differing approaches to evaluate their constitutionality. The Colorado Supreme Court in Seymour took a particularly novel, and perhaps promising, approach and found a possessory interest in search history data based on Google’s Terms of Service (ToS).[260] The Pennsylvania Supreme Court in Kurtz came to the opposite conclusion based on Google’s ToS in an REOP decision.[261] The Georgia Supreme Court assumed without deciding that geofence warrants are a search that they are constitutional.[262] And a plurality of the Texas Court of Criminal Appeals assumed that a geofence warrant was a search and held it to be constitutional.[263]

1.       Colorado Supreme Court: Seymour

In Seymour, the Denver Police investigated a 2020 arson that left five people dead.[264] After two months, the police had no suspects and turned to keyword warrants.[265] Since the police suspected arson, they assumed that the suspects had researched the house or at least used the internet to get directions to the house.[266]

The first reverse search warrant the Denver Police requested from Google was a list of any users who had searched any of the nine variations of the suspected arson address in the fifteen days before the fire.[267] The police also requested “each user’s full name, date of birth, email address, physical address, phone number, and IP address.”[268] Google rejected this warrant, as it did not follow Google’s staged process for reverse search warrants.[269] Likewise, Google rejected the second warrant, in which the police asked for two days of location data for each account.[270]

Google responded to the third warrant, in which the police asked for an “anonymized list” of “‘each device [that searched the address] . . . along with the associated IP address’” in the fifteen days before the fire.[271] Google turned over a spreadsheet of sixty-one searches by eight accounts.[272] The Denver Police found that five of the IP addresses related to the eight accounts were based in Colorado, and they obtained an additional warrant compelling Google to turn over personally identifying account information for the Colorado accounts.[273] Based on this information, the Denver Police arrested Seymour.[274]

Seymour moved to suppress the evidence obtained from Google and argued that the reverse search warrant was unconstitutional because “it was not adequately particularized and lacked individualized probable cause.”[275] The trial court denied the motion.[276] The Colorado Supreme Court, in a wide-ranging opinion, held that Seymour had a “constitutionally protected privacy interest in his Google search history even when revealed only in connection with his IP address and not his name” under the Colorado Constitution.[277] Additionally, it held that Seymour had a “constitutionally protected possessory interest in that same history” under both the Fourth Amendment and Colorado Constitution.[278] The Colorado Supreme Court further held that “Seymour’s Google search history implicates his right to freedom of expression.”[279] The court also held that the keyword warrant did meet the particularity requirement, but the court assumed without deciding that the warrant “required individualized probable cause and that its absence here rendered the warrant constitutionally defective.”[280] Finally, the court relied on the good-faith exception to deny Seymour relief.[281]

The Colorado Supreme Court’s ruling in Seymour is unique in that it held Seymour had a possessory interest in his Google search history.[282] The court conducted both an REOP analysis[283] of Seymour’s Google search history and a property-based analysis.[284] The court’s multistep discussion of REOP from Katz stands in contrast to its straightforward application of property law.[285] The court found that, absent the third-party doctrine, Seymour’s Google search history met both prongs of REOP.[286] Google stores search history in an encrypted database, and Seymour did not try to make his data public, thus satisfying the subjective expectation of privacy requirement of REOP.[287] Furthermore, the court wrote that “[t]he contents of an individual’s internet searches reveal as much, if not in many instances more, about one’s private life than other records in which we’ve previously held individuals have a reasonable expectation of privacy,”[288] meeting the objective expectation of privacy of REOP.

Notwithstanding this analysis, the Colorado Supreme Court concluded that there was a search under the Colorado Constitution but not under the Federal Constitution due to the third-party doctrine.[289] In Seymour, Google responded to the warrant at first by providing only “nominally anonymized IP addresses” that only identified devices and not users.[290] In applying the third-party doctrine, the Colorado Supreme Court held that there was no Fourth Amendment REOP in the search history “when viewed simply in connection with an anonymized IP address” associated with Seymour’s Google account, as he voluntarily shared this information with Google to use its search service.[291]

IP addresses are anything but anonymous:[292] The facts of Seymour illustrate that the police used the “anonymized” IP addresses to determine which devices were in Colorado at the time of the search.[293] The Colorado Supreme Court admitted as much in its opinion when discussing REOP under the Colorado Constitution: “Once law enforcement has an IP address, it can easily associate that IP address with an individual.”[294] By relying on the third-party doctrine and then finding this privacy right in the Colorado Constitution, the court was able to avoid wading too far into the murky waters of REOP and Katz jurisprudence under the Federal Constitution.

The most innovative and interesting part of the Seymour opinion is the court’s “novel” use of property law to find that Seymour had a possessory interest in his Google search history.[295] It is also striking how straightforward the court’s analysis of the property claim is compared to its REOP analysis. Here, the Colorado Supreme Court relied on U.S. Supreme Court precedent to hold that people can have property rights to “intangible property.”[296] The Colorado Supreme Court then concluded that users have a property interest in user content that Google holds. The court cited Google’s own Terms of Service (ToS), which state that “users own their Google content, which . . . includes their search histories.”[297] The court also cited the U.S. Supreme Court case Rakas v. Illinois for the propositions that “[o]ne of the main rights attaching to property is the right to exclude others,” and that the owner of property has a “legitimate expectation of privacy by virtue of this right to exclude.”[298] The Colorado Supreme Court further concluded that since users own their content, which Google stores for them, law enforcement copying of that data is a seizure under the Fourth Amendment as copying “meaningfully interfere[s] with [their] possessory interest in that data.”[299]

The Colorado Supreme Court held that users have a possessory interest in their data kept by Google and a right to exclude other people and thus a privacy right in the data.[300] Therefore, law enforcement copying this data is a seizure under the Fourth Amendment.[301] What is less clear is why the court focused on seizures, rather than searches. As Professor Orin Kerr notes, the concepts of search and seizure in Fourth Amendment law are generally “coextensive.”[302] Here, the court appears to have found that a seizure is a broader category than a search for Fourth Amendment purposes.[303] The court wrote that the government did not search Seymour’s data in accessing it, but did seize it by copying it.[304]

While the Colorado Supreme Court’s analysis is promising, its property-based approach to its Fourth Amendment analysis would be improved if it had held that copying Seymour’s data constituted a search, in addition to a seizure. This Note would follow the dissent’s view that copying Seymour’s data was also a search, since Google had to scan (or search) all users’ records to access and filter Seymour’s digital property.[305] This Note also disagrees with Kerr and proposes that because users have property rights in their data, any government access, not just copying, should constitute a search or seizure under the Fourth Amendment.[306] Kerr appears to argue that copying comes before searching and that copying is only a seizure if the material copied is protected by the Fourth Amendment.[307] Relying on the Colorado Supreme Court’s search analysis, Kerr accepts that Google’s access of Seymour’s data was not a search and thus cannot be seized in the meaning of the Fourth Amendment, since the two are coextensive.[308] In Seymour’s case, this Note argues that the search occurred prior to the copying, when the government compelled Google to produce a list of “anonymized” IP addresses, because Seymour had a privacy interest in his data.[309]

2.       Pennsylvania Supreme Court: Kurtz

In Kurtz, the Pennsylvania Supreme Court also relied on Google’s Terms of Service (ToS) but reached the opposite conclusion as the Colorado Supreme Court: Users do not have a privacy interest in their internet searches, at least those on Google.[310] Kurtz did not assert a property interest in his internet history but relied on a theory of REOP under both the Pennsylvania and Federal Constitutions.[311]

In Kurtz, the Pennsylvania State Police sought a keyword search warrant for the name and address of a sexual assault victim who lived in a remote area of Pennsylvania.[312] Google informed the police that someone had searched the victim’s address twice in the hours before the attack.[313] The Pennsylvania State Police used this lead to arrest and charge Kurtz.[314]

The Pennsylvania Supreme Court’s opinion in Kurtz is in tension with the Colorado Supreme Court’s opinion in Seymour. In Kurtz, the court drew heavily upon Google’s ToS to hold that there is no REOP in keyword search history.[315] The Kurtz court wrote: “Google went beyond subtle indicators. Google expressly informed its users that one should not expect any privacy when using its services.”[316] The court then quoted Google’s Privacy Policy, which includes: “We will share personal information with companies, organizations or individuals outside of Google if we have a good-faith belief that access, use, preservation or disclosure of the information is reasonably necessary to . . . meet any applicable law, regulation, legal process or enforceable request.”[317]

The Pennsylvania Supreme Courtsought to distinguish the facts of Kurtz from the facts of Carpenter.[318] The court held, somewhat bizarrely, that the internet is less pervasive in modern times than cell phones.[319] In an attempt to evade the “involuntariness” aspect of Carpenter, the court wrote: “[T]he use of the internet is not involuntary, as cell phones have become. To the contrary, every time a person logs on to the internet, that person makes a choice.”[320] In a particularly colorful passage, the court justified not finding REOP in search data because physical libraries exist.[321] “Someone hoping to learn more about dinosaurs or galaxies can conduct research in print materials at the library. . . . The point is that the data trail created by using the internet is not involuntary in the same way that the trail created by carrying a cell phone is.”[322]

Kurtz presents an interesting counter to the Colorado Supreme Court’s Seymour decision in that both courts consulted Google’s ToS and came to different conclusions about REOP in internet search history data.

3.       Georgia Supreme Court: Jones

In Jones, the Georgia Supreme Court assumed without deciding that a geofence search warrant is a search.[323] The court further reasoned that such a search is permissible under the Fourth Amendment as long as it meets the probable cause and particularity requirements.[324] This approach to reverse search warrants puts the Georgia Supreme Court somewhere between the gulf of the Fifth Circuit in Smith (search, not permitted by the Fourth Amendment)[325] and the Fourth Circuit panel in Chatrie (not a search).[326]

In its probable cause analysis, the Georgia Supreme Court glossed over the technological aspects of how geofence warrants work.[327] The court characterized Jones’s REOP (not property-based) argument as saying that Google’s search was “overbroad because Google’s initial search for the requested information would ‘touch’ the data of all Google users.”[328] The court then listed several kinds of databases, including physical addresses, phone numbers, license plate numbers, and DNA profiles and stated: “[W]hen an investigator runs on of those searches . . . he sees the results, if any, that the query returns. He does not view every license plate or phone number in the system.”[329]

This reasoning does not work in the reverse search warrant context because it assumes an identified target of the search. In the court’s example, an investigator can use a database to look up a physical address, but she probably has an idea of whose address she is looking up. Contrastingly, in the reverse warrant context, an investigator often has no idea who she is looking for and resorts to a “dragnet” investigation technique.[330] The Georgia Supreme Court, unlike the Colorado Supreme Court, did not address any property rights users may have in their location history data.

4.       Texas Court of Criminal Appeals: Wells

In Wells, the Texas Court of Criminal Appeals took a similar path as the Georgia Supreme Court.[331] The Texas Court of Criminal Appeals assumed without deciding that geofence warrants are searches and held the warrant at issue to be constitutional.[332] The court found that there was probable cause “for which evidence of the perpetrator’s identity could be found in Google’s location history database.”[333]

Relying on the ubiquity of cell phones as a tool of modern life, the Texas Court of Criminal Appeals leapt to the conclusion that probable cause existed because a home invasion perpetrator had a cell phone with Google location history enabled.[334] In Wells, the evidence presented in the warrant affidavit was thinner than that in Chatrie and Smith, as there was nothing in the affidavit suggesting a suspect had a cell phone other than a police officer’s speculation.[335] The Texas Court of Criminal Appeals’s finding of probable cause as to the existence of data in a database is troubling and indicates the tautological nature of reverse warrants: There is always probable cause to believe someone’s data exists in a database of 592 million people.[336] The court ultimately concluded that geofence warrants can meet the probable cause and particularity requirements of the Fourth Amendment.[337]

IV. Property and Reverse Search Warrants

Property is not a panacea for the Fourth Amendment. Indeed, a property-only approach to the Fourth Amendment might erode many of the protections that a REOP framework affords.[338] However, using the protections of property may allow for greater protection of data targeted by reverse search warrants, an area where courts are currently divided under the REOP framework.[339] A property-based approach is additive and meant to expand protection to areas current Fourth Amendment doctrine has difficulty covering. It is not meant to displace the Katz and Carpenter approaches.[340] A property-based approach considers the conservative jurisprudence of the current U.S. Supreme Court and is a pragmatic response to the Court’s dissolution of formerly solid rights in other contexts like abortion.[341]

This Section surveys several main strands of property theory as they relate to the Fourth Amendment and applies useful ideas from each strand to the reverse search warrant context. It begins by examining Justice Gorsuch’s opinion in Carpenter,[342] with a focus on the doctrine of bailment and the right to exclude.[343] It then dives into the nuance of the sources of law courts might draw on for a property-based approach, including positive law,[344] general law,[345] and contracts or Terms of Service (ToS).[346] It also examines what “papers” means within the Fourth Amendment and in relation to reverse search warrants.[347]

Ultimately, this Section suggests embracing the Colorado Supreme Court’s course in Seymour, with several modifications.[348] First, as a baseline, this Note argues that virtually all data targeted by reverse warrants are “papers” under the Fourth Amendment.[349] Next, this Note answers the follow-up question of “whose papers?” by looking to the ToS agreements, statutes, and the doctrine of bailment to help define ownership of the papers and suggests that they belong to the users. Then, this Section asks if a search or seizure occurred and finds that one occurs when the government asks an internet company to query its entire database because it affects the property interests of each individual user.

A.  Existing Approaches to Property and the Fourth Amendment

1.   Property in Carpenter

In Carpenter v. United States, the majority grounded its opinion in a REOP approach to CSLI.[350] However, Justice Gorsuch explored an alternative ground—property and specifically, bailment—in his solo dissent.[351] Scholars have taken Justice Gorsuch’s dissent as an invitation to re-examine property approaches to the Fourth Amendment.[352]

Justice Gorsuch stated in his Carpenter dissent that he was sketching an outline of how the property law concept of bailment could apply to digital data.[353] He wrote: “Much work is needed to revitalize this area . . . I do not begin to have all the answers today, but (unlike with Katz) at least I have a pretty good idea of what the questions are.”[354] As a result, Justice Gorsuch’s opinion is a high-level overview that commentators have criticized as evading much of the technical nitty-gritty of property law or how his approach would work in practice.[355]

Justice Gorsuch wrote: “[T]he traditional approach asked if a house, paper, or effect was yours under law. No more was needed to trigger the Fourth Amendment.”[356] He then identified several principles a property-based approach might consider: (1) third-party possession of papers and effects does not diminish an individual’s interest in them;[357] (2) there is no need for “complete ownership or exclusive control” of property for a Fourth Amendment claim;[358] (3) courts should look to positive law to evaluate new technologies and not fall back on “judicial intuition”;[359] (4) in some circumstances, positive law cannot be used to defeat a Fourth Amendment claim;[360] and (5) the constitutional floor “may, in some instances, bar efforts to circumvent the Fourth Amendment’s protection through the use of subpoenas.”[361]

Most notable in the reverse warrant context are points (1), (2), and (3). In support of point (1), Justice Gorsuch advocates for considering digital data held by third parties to be a bailment.[362] A bailment is a relationship in which the bailor gives their property to the bailee to “hold[] for a certain purpose, usu[ally] under an express or implied-in-fact contract.”[363] The bailor does not give up full ownership of their property, only temporary possession, and the bailee takes care of the property for the bailor.[364] Gorsuch relied on Ex parte Jackson as an example.[365] Ex parte Jackson, a Supreme Court case from the late nineteenth century, found that the government could not open a sealed letter sent through the U.S. Postal Service.[366] In Jackson, the bailor was the writer of the letter, and the bailee was the Postal Service.[367] Although the bailor gave up physical possession of the letter, this did not mean that the contents, which the bailor sought to keep from public view, were vulnerable to the government’s prying eyes.[368]

Courts have applied Ex parte Jackson’sholding to cases involving new technologies. The Sixth Circuit relied on Ex parte Jackson in United States v. Warshak to hold that users have a REOP in emails under Katz.[369] Courts have generally held that the metadata, which includes “to and from” information (much like the addresses on a physical envelope), does not require a warrant, but that the content of the communication does, mirroring the physical distinction present in Ex parte Jackson.[370] Ex parte Jackson was an earlier response to a new technology of the time: mail travelling by Pony Express.[371]

Scholars have expounded upon Justice Gorsuch’s application of bailment to digital data in his Carpenter dissent.[372] Historically, bailments have been physical property,[373] like “entrusting a horse to an innkeeper,” “parking a car in a secure garage,” or “shipping.”[374] However, bailment can apply to digital files held by a third-party without much of a leap.[375] Under a “bundle of rights” approach to property rights, the various rights of ownership (right to create, possess, destroy, use, exclude, etc.) can be thought of as distinct rights comprising ownership.[376] A bailor does not give up all of the rights in the bundle when giving her property to a bailee.[377] Crucially, the bailor retains the right to exclude—a fact the Colorado Supreme Court in Seymour viewed as significant when it found that users have a property interest in their search history data held by Google.[378]

Indeed, as Professor Laura Donohue argues, historically “[t]he bailee had possession, while the bailor retained residual ownership rights.”[379] Donohue outlines several kinds of bailment that existed at common law, including “use of the thing,” “work on a thing,” “the keeping of a thing,” and “the transportation of an item,” and envisions how they would apply to the CSLI at issue in Carpenter.[380] Donohue argues that CSLI fits neatly into the “use of a thing” category, as “telecommunication service providers only have access to customers’ location while they are paying for the service.”[381] Applying this logic to the data targeted by reverse warrants is not an insurmountable task. Location data that Google collects is functionally the same as CSLI, but even more precise.[382] Likewise, the Seymour court held that users, not Google, own their search history, even if Google may use it to provide services.[383] Even critics of a property-based approach to the Fourth Amendment, like Professors Matthew Tokson and Michael Pollack, acknowledge that contexts like “digital document storage or email content . . . could be analogized to a bailment relationship with the service provider.”[384]

2.   Positive and General Law

Under a property law framework, where should judges look to guide their search and seizure analyses? In Carpenter, Justice Gorsuch proposed that judges should look to “positive law” for guidance when applying a property-based approach to the Fourth Amendment.[385]

Positive law, as articulated in Professors William Baude and James Y. Stern’s influential article, The Positive Law Model of the Fourth Amendment, asks if the government has “engaged in an investigative act that would be unlawful for a similarly situated private actor to perform.”[386] In other words, courts assess whether the government action violated legislatively enacted statutes like “property law, privacy torts, consumer law . . . and other provision[s] . . . rather than a freestanding doctrine of privacy fashioned by courts on the fly.”[387] Baude and Stern argue that their approach “is clearer, more predictable, more accommodating of variation in different times and places, and more sensitive to the institutional strengths of legislative bodies, particularly when it comes to issues presented by new technologies.”[388] Justice Gorsuch suggested that judges should look to federal and state laws, like the federal Stored Communications Act, for guidance in determining what constitutes digital property and whether government access to it is a search or seizure.[389] Under a positive law approach, Fourth Amendment violations are tied to statutory law violations.[390]

A positive law framework could leave a patchwork of Fourth Amendment protections, dependent on legislatures and jurisdiction-specific laws.[391] Seeking to address these shortcomings, Professors Danielle D’Onfro and Daniel Epps propose a property-based approach to the Fourth Amendment based on “general law.”[392] General law, as used by D’Onfro and Epps, is a wide-ranging concept defined as “the common law under the control of no particular sovereign.”[393] General law allows for an expansive inquiry into the common law as it has developed over time and allows judges to consider “custom, tradition, and social facts.”[394] While Baude and Sterne’s positive law approach to property, limited to statutes passed by legislatures, may be narrow, scholars have criticized D’Onfro and Epps’s general law approach as impossibly wide.[395] Judges would have latitude to consider a great deal of material under a general law approach.[396] Additionally, the approach seems to replicate some of the key issues with the Katz test, namely that allowing judges to inquire into “social facts” results in an unclear and subjective framework.[397] Professor Maureen Brady cautions, “[g]iven the wide and varied sources of general law and the vagaries of common-law standards, judges could certainly find justifications in the ‘general law’ under which to smuggle their own views of the proper scope of a person’s Fourth Amendment rights.”[398] While general law is appealing because it can account for new situations and technologies in light of changing social facts, the sheer scope of material to draw on may make it unpredictable and difficult to apply.

3.   Contracts and Terms of Service

In addition to positive law and general law, courts[399] and scholars[400] have looked to contracts and Terms of Service (ToS) to construe Fourth Amendment rights. Most prominently, the Colorado Supreme Court in Seymour identified a property interest in Google users’ search history data based on the language of Google’s ToS.[401] This approach emphasized the right to exclude inherent in the users’ possessory interest in the data as made clear by Google’s ToS.[402] On the other hand, the Pennsylvania Supreme Court in Kurtz relied on Google’s ToS to hold that users have no REOP in their search history because the ToS informs them of what data Google collects and that it will share this data in compliance with “any applicable law, regulation, legal process or enforceable request.”[403] This gulf between two state supreme courts on the same reverse keyword warrant issue demonstrates the unsettled nature of the law and indicates that courts look to ToS to understand how to approach issues of reverse warrants.[404]

Professor Orin Kerr argues that ToS should not reduce Fourth Amendment rights because they are contracts between a user and a private company, not the government.[405] Kerr cautions that ToS can “nullify”[406] the Fourth Amendment and argues that courts that have relied on ToS to reduce Fourth Amendment protections are mistaken.[407] Kerr explains the concept of ToS “nullifying” Fourth Amendment rights as follows: “Fourth Amendment protection requires rights; [ToS] define rights in online accounts; and therefore Terms of Service define Fourth Amendment protection in online accounts.”[408] Kerr argues that courts are incorrect to define Fourth Amendment rights in online accounts based on what internet companies specify in ToS, because ToS are not agreements between the user and the government, but the user and the internet company.[409]

But what about when courts, like the Colorado Supreme Court in Seymour, use ToS to reinforce Fourth Amendment rights?[410] In his article, Terms of Service and Fourth Amendment Rights, Kerr focuses on whether ToS can reduce Fourth Amendment rights.[411] He has written in commentary on the Seymour and Kurtz decisions that “the reliance on [ToS] [is] a crutch both way. . . . I don’t think [ToS] alter Fourth Amendment rights—up or down.”[412]

Student notes about courts’ use of ToS to find property interests are divided. For example, in Finding Rights in the Fine Print, Dalia Wrocherinsky argues that the Colorado Supreme Court got it right in Seymour and that search history can be property based on Google’s ToS.[413] Wrocherinsky writes that the property interest in search history flows in part from Google’s ToS but also from the Colorado Privacy Act.[414] The rights Coloradans receive from this act “implicate the critical property rights to exclude and manage, strengthening the argument that Colorado consumers have property rights in their search history.”[415] While the Colorado Supreme Court addressed the digital privacy protections of the Stored Communications Act and the Colorado Privacy Act, it did so in the REOP analysis section of its opinion and not in the property section.[416] On the other hand, Alec J.H. Block and Joseph W. Paul argue that only positive law, not ToS, can create property rights.[417] They explain that finding a property right that is good against the world from a contract or ToS, which is only good between the parties, is untenable.[418]

Regardless of these scholars’ views, the Seymour court explicitly relied on Google’s ToS to find a property interest under both the Colorado and Federal Constitutions and, in doing so, established it as the law in Colorado.[419] The court’s decision in Seymour strikes a blow against the third-party doctrine and is a small step in reaching equity in the balance between massive internet companies and individual criminal defendants.[420] However, as the court in Kurtz demonstrated, ToS are subject to varying interpretations and may not always increase Fourth Amendment protections. Nevertheless, even if ToS cannot create property rights, they at least provide a useful guide for courts to consider in their analysis of a property-based approach to the Fourth Amendment.[421] In a world where the U.S. Supreme Court seems intent on eroding existing rights and protections,[422] and several members have made their disapproval of Katz[423] known, advocates should consider every tool they have to preserve constitutional rights.[424]

4.   Data as “Papers?”

Outside of purely doctrinal property approaches, “property-inspired”[425] approaches ask if the data targeted by a reverse warrant is the contemporary equivalent of “papers” as enshrined in the Fourth Amendment.[426] If data qualifies as papers, then it deserves Fourth Amendment protection.[427] Justice Gorsuch in Carpenter stated that data, in particular CSLI, can constitute papers within the meaning of the Fourth Amendment.[428] Justice Kennedy, in his dissent joined by Justices Alito and Thomas, wrote that the data at issue in Carpenter was not Carpenter’s papers but acknowledged that “Miller and Smith may not apply when the Government obtains the modern-day equivalents of an individual’s own ‘papers’ or ‘effects,’ even when those papers or effects are held by a third party” and that this relationship would be a bailment.[429] Thus, it appears that the current composition of the U.S. Supreme Court may be receptive to a reverse warrant argument that characterizes targeted data as the “modern-day equivalents” of “papers.”[430]

What counts as an individual’s papers? Michael W. Price argues that papers should be anything with expressive or associational value and cites the history of the Fourth Amendment bolstering the First Amendment for this theory.[431] Price argues that “personal files stored in the cloud, as well as communications data and metadata” qualify as “papers.”[432] He relies on the U.S. Supreme Court’s history of expansively interpreting “houses” within the Fourth Amendment’s text to claim that “papers” should include digital equivalents.[433] Price goes further and states that metadata should also receive the protection of Fourth Amendment papers.[434] This is a direct strike against the third-party doctrine, as Price notes: “The text of the Fourth Amendment, however, does not actually draw a distinction between one’s private papers and information about those papers, between data and metadata. Viewed another way, metadata is just additional private data that happens to be stored by a third party.”[435] Rooting protections in the text of the Fourth Amendment has appeal to the conservative U.S. Supreme Court.[436]

In a similar vein, Professor Laura Donohue advocates for a “but for” approach to determine who owns digital data.[437] Data ownership is a thorny issue, and Donohue’s approach attempts to avoid wading into the murky doctrinal waters of property ownership.[438] Donohue argues that a “but for” approach evaluating CSLI would focus on the fact that “the right-holder buys the phone, charges it, turns it on, and decides when to carry it. It is up to the right-holder to decide with whom the resulting information is shared.”[439] Thus CSLI, which is held by a third-party network provider, only materializes due to the right-holders’ actions.[440] Generating data like CSLI is “functionally equivalent” to an individual’s decisions in “writing a letter, keeping a diary, dictating a memo, engaging in business transactions, and going about one’s daily business.”[441] The person who created (or whose actions caused the creation of) the data and who has the legal right to determine whether and with whom it is shared “hold[s] an ownership interest” in that data.[442] Donohue then argues that the best way to conceive of the relationship between a user, the data they generate, and an internet company is through bailment and relies on Justice Gorsuch’s bailment sketch in Carpenter.[443]

B.  Applying Property Protections to Reverse Search Warrant Data

Applying a property-based approach to the constitutionality of reverse search warrants requires a three-part analysis. First, are the data that reverse search warrants target “papers” under the Fourth Amendment? Second, if so, do property-like protections apply to a user’s data or papers? And third, when does a search or seizure of the data or papers occur? To answer these questions, this Section combines useful aspects of the property frameworks discussed previously.

The first question is whether the data targeted are “papers” under the Fourth Amendment. Using Price’s idea that if data has expressive or associational value, both geofence and keyword search warrant data clear this hurdle.[444] Geofence data is location data, and its comprehensive nature can reveal a person’s associations. As Google wrote in its amicus brief in Chatrie, quoting the Court in Carpenter, “Google [location history] information ‘provides an intimate window into a person’s life, revealing not only his particular movements, but through them, his familial, political, professional, religious, and sexual associations.’”[445] Likewise, keyword search data goes to the heart of expressive data because it consists of a person’s thoughts and questions typed into a search bar.[446] As the Colorado Supreme Court held in Seymour: “[T]he act of accessing a webpage to receive information—just like purchasing a book—constitutes an expressive activity because the user is seeking to tap into the marketplace of ideas.”[447] Thus, asking whether data has an expressive or associational value, and therefore qualifies as “papers” under the Fourth Amendment, simplifies the first step of the search/seizure analysis.

Next, the analysis turns on whether the papers belong to the user and not to the internet company on whose servers the papers reside. Again, the answer is yes for both geofence and keyword search warrants. The responses of scholars and courts to Justice Gorsuch’s opinion sketching an idea of digital bailment in Carpenter demonstrate that users maintain a property interest in their data under the principles of bailment and positive or general law,[448] terms of service (ToS) agreements,[449] and “property-inspired” approaches like Donohue’s “but for” approach.[450] Courts should follow the Colorado Supreme Court in Seymour and look to ToS agreements as guidelines, as well as rely on positive law like state privacy statutes to determine what property-based protections apply to user data.[451] As the Colorado Supreme Court held, users maintain a possessory interest in their data through the right to exclude others.[452] That internet companies possess user data and use it for their own operations does not strip users of all of their property rights—a conclusion supported by the historical doctrine of bailment,[453] ToS agreements,[454] and state and federal statutes.[455]

Finally, the inquiry asks if a search or seizure occurred. Under Katz, this is a complex calculus depending on an individual’s subjective expectation of privacy and whether society recognizes that expectation as reasonable,[456] but it is relatively simple in the reverse search warrant context.

As Justice Gorsuch wrote in Carpenter, the property-based analysis looks to whether “a house, paper, or effect was yours under law. No more was needed to trigger the Fourth Amendment.”[457] The first two steps of this analysis, asking if the data qualifies as papers and if the papers belong to the user, lead to the conclusion that a search likely occurs at step one of the reverse warrant process. Step one is when “law enforcement receives a warrant compelling Google to disclose a de-identified list of all Google users whose Location History data indicates were within the geofence during a specified timeframe.”[458] Step one requires Google to “search all location history data to identify users whose devices were present within the geofence during the defined timeframe.”[459] Under a property-based theory of the Fourth Amendment, because individual users have property interests in their data, the government’s attempt to access that data constitutes a search or seizure. Finding that a search occurred at step one puts this approach in alignment with the Fifth Circuit’s holding in Smith.[460] As the court explained, it is “constitutionally insufficient” to make Google, or another internet company, “search through the entirety of its [database]—all 592 million accounts—for all of their locations at a given point in time” while “law enforcement officials have no idea who they are looking for, or whether the search will turn up a result.”[461] The property-based approach attempts to provide maximum protection for user data while avoiding a complex analysis to determine at what step of the reverse search warrant process a REOP search occurs.

Conclusion

A property-based approach to reverse search warrants strengthens the protections of the Fourth Amendment and simplifies complex cases. In the context of reverse warrants, courts have an opportunity to do something unique: use an old doctrine to solve current and future problems related to emerging technologies. Reverse warrants take advantage of the scale and computing capacity of cloud-based internet companies to mine user data for criminal investigative purposes.[462] Whether geofence or keyword warrants, both forms of technology are searches under the Fourth Amendment and fail to comply with the Fourth Amendment’s probable cause and particularity requirements. This rummaging poses an immense threat to the privacy and data security of all users. As one court explained in denying an application for a geofence warrant, “[Law enforcement] is essentially asking the Court to allow it access to an entire haystack because it may contain a needle.”[463] The absurdity of reverse warrants comes into stark relief when one realizes that the “haystack” the government asks to search could be the five trillion annual Google searches[464] or the location information of 592 million Google users.[465]

Some courts have twisted themselves (and the Katz reasonable expectation of privacy doctrine) into knots to find that reverse warrants are not searches.[466] This has led to a split among the Fourth and Fifth Circuits, where a Fourth Circuit panel ruled that geofence warrants are not searches (later vacated by a divided en banc court) while the Fifth Circuit held that geofence warrants are searches and that they are unconstitutional general warrants.[467] State courts are also split, with the Georgia Supreme Court[468] and Texas Court of Criminal Appeals[469] holding that geofence warrants are constitutionally permissible searches, while the Colorado Supreme Court found keyword searches to be acceptable in the specific case of Seymour but left open the possibility of revisiting reverse search warrants.[470] The split is most acute in Texas, where the Fifth Circuit banned the use of geofence warrants in federal court,[471] while the Court of Criminal Appeals upheld their legality in state court.[472] The source of much of this tension is how courts have applied the Katz test to determine if users have a “reasonable expectation of privacy” in their location data held by Google.[473] Such a split will soon be adjudicated by the U.S. Supreme Court in Chatrie.

This Note has proposed a property-based conception of privacy as a framework to resolve the current split on reverse search warrants. Instead of asking the doctrinally difficult question of “does the affected person have a reasonable expectation of privacy interest” in the data the government accessed, a property-based conception asks, does the user data qualify as “papers” under the Fourth Amendment, and does the user have an ownership interest in it? If the user does have an ownership interest, the government must obtain either a warrant or the user’s consent to search or seize the data. Reverse search warrants can never satisfy the probable cause and particularity requirements of the Fourth Amendment as by their very design they are massive, indiscriminate dragnets that leave far much discretion with the individual officers executing the warrant. Thus, in absence of legislative action banning reverse search warrants, courts around the country should supplement the Katz approach with a property-based conception to find reverse warrants to be unconstitutional.

In announcing its Seymour decision, the Colorado Supreme Court wrote: “If dystopian problems emerge, as some fear, the courts stand ready to hear argument regarding how we should rein in law enforcement’s use of rapidly advancing technology.”[474] The age of “dystopian problems” is here. Advocates and courts can fight against authoritarianism and strengthen the Constitution by shutting down reverse search warrant using tools from an age-old doctrine: property.

Copyright © 2026 Nicholas Delehanty, J.D., 2026, University of California, Berkeley, School of Law. Thank you toProfessor Eric Stover and the members of the 2025 Social Justice and Human Rights Writing Seminar for their helpful feedback, especially Kaylana Mueller-Hsia and Katarina van Alebeek. I am also grateful to the many California Law Review editors who had a hand in making this Note as good as it can be, particularly Samahria Alpern who improved all aspects of it. Thank you to William Weightman for clarifying my thinking on the complexities of property. And, of course, thank you to Olivia Craighead for everything, including watching all of the cited movies with me. All views and mistakes are my own.

[1].‍ ‍Jon Kelly, The Mysterious World of Safe Deposit Boxes, BBC News (Apr. 9, 2015), https://www.bbc.com/news/magazine-32223768 [https://perma.cc/KGE7-CK52] (describing safe deposit boxes as “useful plot device[s]”).

       [2]. See, e.g., Thief (United Artists 1981) (demonstrating characters’ high degree of skill in breaking into a bank vault to steal diamonds from safe deposit boxes); Casino (Universal Pictures 1995) (character puts $2 million in cash in safe deposit box and gives key to untrustworthy associate); The Bourne Identity (Universal Pictures 2002) (illustrating the impenetrability of safe deposit box when former CIA agent retrieves passports, cash, and handgun from one); Inside Man (Universal Pictures 2006) (entirety of film is about elaborate heist to break into a single safe deposit box).

       [3]. Thief, supra note 2.

       [4]. The Bourne Identity, supra note 2.

       [5]. Casino, supra note 2.

       [6]. Ben Eisen & Shara Tibken, No One Can Find Safe-Deposit Boxes Anymore, Wall St. J. (Nov. 15, 2024), https://www.wsj.com/personal-finance/savings/no-one-can-find-safe-deposit-boxes-anymore-0bfccf88 [https://perma.cc/9N2B-5JJF].

       [7]. See, e.g., Esther van Laarhoven-Smits, If Data’s the New Gold, How Do You Use It to Create Value?, EY (Apr. 6, 2025),  https://www.ey.com/en_ch/insights/big-data-analytics/if-datas-the-new-gold-how-do-you-use-it-to-create-value [https://perma.cc/FY9E-BYBN]; Sandro Shubladze, How To Make Use of the New Gold: Data, Forbes (Mar. 27, 2023), https://www.forbes.com/councils/forbestechcouncil/2023/03/27/how-to-make-use-of-the-new-gold-data/ [https://perma.cc/U8KA-VL78]; The World’s Most Valuable Resource Is No Longer Oil, But Data, Economist (May 6, 2017), https://www.economist.com/leaders/2017/05/06/the-worlds-most-valuable-resource-is-no-longer-oil-but-data [https://perma.cc/9L5Z-NXR8].

[8]. See Danielle D’Onfro, The New Bailments, 97 Wash. L. Rev. 97, 102 (2022) (arguing that “[c]loud storage is the safe deposit box of the twenty-first century, and upon closer inspection, the law of safe deposit boxes maps nicely onto cloud storage”); see also Jacob S. Sherkow, Natalie Ram & Carl A. Gunter, Familial Searches, the Fourth Amendment, and Genomic Control, 96 S. Cal. L. Rev. Postscript 1, 11 (2022) (“Using a bank as an analogy—with an admission that all analogies are imperfect—genomic data would, at most, be like a safe deposit box to which the possibility of access means neither knowledge nor control of its contents. And, generally, warrants are required for safe deposit boxes.”).

       [9]. 565 U.S. 400, 417 (2012) (Sotomayor, J., concurring).

     [10]. Jonathan Askin, From Data Co-Opting to Data Co-Oping: Using New Corporate Structures, Data Corporate-Hood, and Data Personhood to Prioritize Data Privacy, 20 U.C.L. Bus. J. 177, 206 (2024).

     [11]. See Jones, 565 U.S. at 417 (discussing users sharing information about phone numbers they text and URLs they visit with the companies that provide these services).

[12]. See Mary D. Fan, Big Data Searches and the Future of Criminal Procedure, 102 Tex. L. Rev. 877, 934 (2024) (“The massive vault of data that companies have about our search histories, movements, and behaviors is giving rise to new police investigative strategies.”); Ronald J. Rychlak, Geofence Warrants: The New Boundaries, 93 Miss. L.J. 957, 962 (2024) (arguing that new technologies, like geofence warrants, are “valuable tool[s]” for law enforcement).

     [13]. See Oral Argument at 02:33, United States v. Chatrie, 136 F.4th 100 (4th Cir. 2025) (No. 22-4489) (on file with the California Law Review).

     [14]. Fourth Amend. Ctr., Nat’l Ass’n of Crim. Def. Laws., Geofence Warrant Primer 1 (2022).

[15]. See United States v. Smith, 110 F.4th 817, 823–24 (5th Cir. 2024) (claiming that a third of Google users or 592 million people had location history enabled in 2018).

     [16]. See United States v. Chatrie, 590 F. Supp. 3d 901, 908 (E.D. Va. 2022) (“Both McGriff and Rodriguez declared that, to identify users within the relevant timeframe of a geofence, Google has to compare all the data in the Sensorvault in order to identify users within the relevant timeframe of a geofence.”), aff’d, 107 F.4th 319 (4th Cir. 2024), aff’d on reh’g en banc, 136 F.4th 100 (4th Cir. 2025), cert. granted in part, 223 L. Ed. 2d 553 (U.S. 2026).

     [17]. United States v. Chatrie, 107 F.4th 319, 325 (4th Cir. 2024), aff’d on reh’g en banc, 136 F.4th 100 (4th Cir. 2025), cert. granted in part, 223 L. Ed. 2d 553 (U.S. 2026). In addition to geofence warrants, reverse search warrants also include keyword warrants, in which the government asks a search engine or other data provider to turn over all user information for certain keywords entered into the search engine.

     [18]. See, e.g., Note, Geofence Warrants and The Fourth Amendment, 134 Harv. L. Rev. 2508 (2021); Danny Drane, Note, Why It’s Time to Ban Geofence Searches in Light of United States v. Chatrie, 38 Berkeley Tech. L.J. 1307 (2023); Orin Kerr, The ACLU’s Response to My Post on the Fifth Circuit’s Smith Ruling—And My Reply to the ACLU, Reason: Volokh Conspiracy (Aug. 16, 2024), https://reason.com/volokh/2024/08/16/the-aclus-response-to-my-post-on-the-fifth-circuits-smith-ruling-and-my-reply-to-the-aclu/ [https://perma.cc/4UAE-HQKT]; Robert Frommer, How To Use Geofence Warrants in a Constitutional Manner, Law360 Persps. (Sep. 6, 2024), https://www.law360.com/articles/1876930/how-to-use-geofence-warrants-in-a-constitutional-manner [https://perma.cc/7X27-JENB]; Jackie O’Neil, Much Ado About Geofence Warrants, Harv. L. Rev. Blog (2025), https://harvardlawreview.org/blog/2025/02/much-ado-about-geofence-warrants/ [https://perma.cc/4U5A-C4SP]; Jennifer Lynch, Is This the End of Geofence Warrants?, Elec. Frontier Found. (Dec. 13, 2023), https://www.eff.org/deeplinks/2023/12/end-geofence-warrants [https://perma.cc/ZQT5-KFJ9]. Reverse search warrants are also the subject of much popular press attention. See, e.g., Raksha Vasudevan, 3 Teens Almost Got Away with Murder. Then Police Found Their Google Searches, Wired (May 21, 2025), https://www.wired.com/story/find-my-iphone-arson-case/ [https://perma.cc/7LX2-KS4F].

[19]. Geofence Warrants and the Grand Canyon of Circuit Splits, Project for Priv. & Surveillance Accountability (Aug. 20, 2024), http://www.protectprivacynow.org/1/post/2024/08/geofence-warrants-and-the-grand-canyon-of-circuit-splits.html [https://perma.cc/RQ6N-JBZM].

[20].‍ ‍107 F.4th 319 (4th Cir. 2024), aff’d on reh’g en banc, 136 F.4th 100 (4th Cir. 2025), cert. granted in part, 223 L. Ed. 2d 553 (U.S. 2026).

[21]. CompareChatrie, 107 F.4th at 339 (holding that “the government did not conduct a Fourth Amendment search when it accessed two hours’ worth of Chatrie’s location information that he voluntarily exposed to Google”), with United States v. Smith, 110 F.4th 817, 840 (5th Cir. 2024) (“We hold that geofence warrants are modern-day general warrants and are unconstitutional under the Fourth Amendment.”).

[22]. See United States v. Chatrie, 136 F.4th 100 (4th Cir. 2025), cert. granted in part, 223 L. Ed. 2d 553 (U.S. 2026).

     [23]. See id.

[24].‍ ‍Chatrie v. United States, 223 L. Ed. 2d 553 (U.S. 2026).

[25]. See, e.g., Jones v. State, 913 S.E.2d 700 (Ga. 2025) (finding that a geofence search was constitutional); People v. Seymour, 536 P.3d 1260 (Colo. 2023) (holding a keyword search to be constitutional, but finding that users have both privacy and possessory interests in their search history); Wells v. State, 714 S.W.3d 614, 616 (Tex. Crim. App.), reh’g denied, 721 S.W.3d 260 (Tex. Crim. App. 2025), petition for cert. filed, No. 25-484 (U.S. Oct. 16, 2025) (holding geofence warrant established probable cause and particularity); Commonwealth v. Kurtz, 348 A.3d 133, 156 (Pa. 2025) (finding no reasonable expectation of privacy in Google search history). The Minnesota Supreme Court recently held that geofence warrants are not unconstitutional “general warrants.” See State v. Contreras-Sanchez, 5 N.W.3d 151 (Minn. Ct. App. 2024), rev’d,No. A22-1579, 2026 WL 1015919, at *12 (Minn. Apr. 15, 2026).

[26]. See sources cited supra note 18.

     [27]. The only federal or state supreme court to address a property-based argument in the reverse search warrant context so far is the Colorado Supreme Court, and it did so briefly in Seymour. People v. Seymour, 536 P.3d 1260, 1273 (Colo. 2023). However, Judge Richardson did address the property-based argument in his opinion in Chatrie and his concurrence in the en banc rehearing of Chatrie. In Chatrie, Judge Richardson found that Chatrie had forfeited his property-based argument by not mentioning it in his opening brief, and in reaching the merits, he faulted Chatrie for not citing positive law and for writing that relying on Google’s terms of service agreement is “an incredibly thin reed on which to hang such a bold pronouncement.” United States v. Chatrie, 107 F.4th 319, 332 n.20 (4th Cir. 2024), aff’d on reh’g en banc, 136 F.4th 100 (4th Cir. 2025), cert. granted in part, 223 L. Ed. 2d 553 (U.S. 2026). Judge Richardson did not mention the forfeiture issue in the en banc concurrence and virtually copy-and-pasted the footnote about the merits of the property claim from his earlier opinion. See United States v. Chatrie, 136 F.4th 100, 141 n.21 (4th Cir. 2025), cert. granted in part, 223 L. Ed. 2d 553 (U.S. 2026).

[28]. See Katz v. United States, 389 U.S. 347, 361 (1967) (Harlan, J., concurring).

     [29]. But see Reed Sawyers, For Geofences: An Originalist Approach to the Fourth Amendment, 29 Geo. Mason L. Rev. 787, 796–806 (2022) (analyzing geofence warrants from an originalist and property-based conception of the Fourth Amendment and concluding that they are not a search but instead a subpoena); Alec J.H. Block & Joseph W. Paul, Fourth Amendment Trespass and Internet Search History, 11 Va. L. Rev. Online 188 (2025) (discussing the application of property-based conception of the Fourth Amendment to keyword search warrants).

[30]. See Defendant Okello T. Chatrie’s Motion to Suppress Evidence Obtained from a “Geofence” General Warrant at 14–16, United States v. Chatrie, 590 F. Supp. 3d 901 (E.D. Va. 2022) (No. 3:19cr130) (Part I.B.2.C entitled “The Acquisition of Defendant’s GPS Data from Google Was a Search Under a Property-Based Approach to the Fourth Amendment.”); Motion to Suppress Evidence from a Keyword Warrant & Request for a Veracity Hearing at 15, People v. Seymour, No. 21CR20001, 2022 WL 21817804 (Colo. Dist. Ct. June 30, 2022) (Motion to suppress evidence brief with Argument Part I.C entitled “This is a search because users have a possessory interest in their keyword search data.”).

[31]. See generally 138 S. Ct. 2206, 2261–72 (2018) (Gorsuch, J., dissenting).

     [32]. SeeChatrie, 107 F.4th at 332 n.20 (deeming property-based argument to be forfeited).

     [33]. In the 2025 term, there were four major geofence cases appealed to the U.S. Supreme Court, Chatrie, Smith, Davis,and Wells. The Court granted cert in Chatrie, Chatrie v. United States, 223 L. Ed. 2d 553 (U.S. 2026), denied cert in Smith in November 2025, Smith v. United States, 146 S. Ct. 356 (2025), and Davis and Wells are still pending as of March 2026, United States v. Davis, 109 F.4th 1320 (11th Cir. 2024), petition for cert. filed, No. 25-5189 (U.S. July 24, 2025); Wells v. State, 714 S.W.3d 614, 616 (Tex. Crim. App.), reh’g denied, 721 S.W.3d 260 (Tex. Crim. App. 2025), petition for cert. filed, No. 25-484 (U.S. Oct. 16, 2025). Comparatively, Davis has less analysis of reverse warrant issues than Chatrie or Wells because the Eleventh Circuit affirmed the district court’s finding that Davis lacked Fourth Amendment standing to challenge the geofence warrant in his case. Davis, 109 F.4th at 1331.

     [34]. This narrow focus on data targeted by reverse search warrants takes into mind Professors Matthew Tokson and Michael Pollack’s and Professor Maureen Brady’s critiques of property frameworks as complex, replicating the Katz framework, and too broad. See generally Michael Pollack & Matthew Tokson, Decentering Property in Fourth Amendment Law, 92 U. Chi. L. Rev. 705 (2024); Maureen E. Brady, The Illusory Promise of General Property Law, 132 Yale L.J.F. 1010 (2023).

     [35]. See, e.g., William Baude & James Y. Stern, The Positive Law Model of the Fourth Amendment, 129 Harv. L. Rev. 1821, 1836 (2016) (analyzing the “cases for the positive law model”).

     [36]. See generally Danielle D’Onfro & Daniel Epps, The Fourth Amendment and General Law, 132 Yale L.J. 910 (2023).

     [37]. See generally Dalia Wrocherinsky, Finding Rights in the Fine Print: How Terms of Services Agreements Can Turn Consumer Search History into Digital Property, 14 Am. U. Bus. L. Rev. 501 (2024).

     [38]. See, e.g., Pollack & Tokson, supra note 34, at 708–10 (critiquing a property-based approach for being too complex and less effective than a privacy approach); Brady, supra note 34, at 1014–15 (critiquing existing property-based approaches for complexity and failure to align with property law); D’Onfro & Epps, supra note 36, at 972–80 (discussing bailment as it applies to the Fourth Amendment); Laura Donohue, Functional Equivalence and Residual Rights Post-Carpenter: Framing a Test Consistent with Precedent and Original Meaning, 2018 S. Ct. Rev. 347, 409 (2018) (advocating for a property-based approach to digital data using a “but for” test of user data creation).

     [39]. See United States v. Chatrie, 107 F.4th 319 (4th Cir. 2024), aff’d on reh’g en banc, 136 F.4th 100 (4th Cir. 2025), cert. granted in part, 223 L. Ed. 2d 553 (U.S. 2026).

     [40]. See United States v. Smith, 110 F.4th 817 (5th Cir. 2024).

     [41]. See cases cited supra note 25.

[42]. See People v. Seymour, 536 P.3d 1260, 1273 (Colo. 2023).

     [43]. Also of interest, but beyond the scope of this Note, is whether reverse search warrants may ever comply with the Fourth Amendment warrant requirements of particularity and probable cause. This analysis is not all that different under a property-based approach, and the Fifth Circuit’s REOP approach in Smith, finding that geofence warrants are unconstitutional general warrants, is persuasive. While reverse search warrants may be useful investigatory devices for law enforcement, this Note takes the position that they are not constitutional as they cannot satisfy the probable cause and particularity requirements. See, e.g., Haley Amster & Brett Diehl, Against Geofences, 74 Stan. L. Rev. 385, 433–35 (2022); Geofence Warrants and The Fourth Amendment, supra note 18, at 2515; Smith, 110 F.4th at 837.

     [44]. See infra Part II.B.1.

     [45]. See infra Parts II.B.3, IV.B.

[46].‍ ‍U.S. Const. amend. IV; see Cong. Rsch. Serv., Modern Doctrine on Selective Incorporation of Bill of Rights, Const. Annotated: Libr. of Cong., https://constitution.congress.gov/browse/essay/amdt14-S1-4-3/ALDE_00013746/ [https://perma.cc/Y8YV-FUJA].

[47]. See Illinois v. Gates, 462 U.S. 213, 238 (1983) (setting out a “totality-of-the-circumstances” standard for probable cause determinations in which the judicial officer asks whether “there is a fair probability that contraband or other evidence of a crime will be found in a particular place”).

[48]. See Johnson v. United States, 333 U.S. 10, 14 (1948) (holding that a “neutral and detached magistrate” and not an officer should make determination to issue a warrant, as it provides a check on the power of the police).

[49]. See Marron v. United States, 275 U.S. 192, 196 (1927) (“The requirement that warrants shall particularly describe the things to be seized makes general searches under them impossible and prevents the seizure of one thing under a warrant describing another.”).

[50]. See Matthew Guariglia, Geofence Warrants and Reverse Keyword Warrants Are So Invasive, Even Big Tech Wants to Ban Them, Elec. Frontier Found. (May 13, 2022), https://www.eff.org/deeplinks/2022/05/geofence-warrants-and-reverse-keyword-warrants-are-so-invasive-even-big-tech-wants [https://perma.cc/8JDB-KSCL] (describing typical warrant process and contrasting it with reverse warrant process).

[51]. Id.; see also Daniel J. Solove, A Taxonomy of Privacy, 154 U. Pa. L. Rev. 477, 507 (2006) (describing that when many disparate pieces of data are aggregated, it can reveal more information than each piece of data on its own); Emily Berman, When Database Queries Are Fourth Amendment Searches, 102 Minn. L. Rev. 577, 577–78 (2017) (noting that the volume of information the government and private actors collect is becoming “ever larger” and analytical tools used “ever more powerful” in revealing new insights in large datasets and suggesting that database queries that aggregate information in this way should be subject to Fourth Amendment restrictions).

     [52]. Jennifer Valentino-DeVries, Tracking Phones, Google Is a Dragnet for the Police, N.Y. Times (Apr. 13, 2019), https://www.nytimes.com/interactive/2019/04/13/us/google-location-tracking-police.html [https://perma.cc/MBW6-YLFW].

[53]. Reverse Search Warrants, Nat’l Ass’n of Crim. Def. Laws., https://www.nacdl.org/Landing/Reverse-Search-Warrants [https://perma.cc/QF62-YHJ5].

[54]. Id.

     [55]. Lynch, supra note 18.

     [56]. See People v. Seymour, 536 P.3d 1260, 1268 (Colo. 2023).

[57]. Reverse Search Warrants, supra note 53.

[58].‍ ‍United States v. Smith, 110 F.4th 817, 823, 837–38 (5th Cir. 2024).

     [59]. See Lynch, supra note 18.

     [60]. See Google, Supplemental Information on Geofence Warrant Requests in the United States, https://services.google.com/fh/files/misc/supplemental_information_geofence_warrants_united_states.pdf [https://perma.cc/Z9W2-SHRA].

     [61]. See generally Marlo McGriff, Updates to Location History and New Controls Coming Soon to Maps, Google (Dec. 12, 2023), https://blog.google/products/maps/updates-to-location-history-and-new-controls-coming-soon-to-maps/ [https://perma.cc/9Y83-7KW7].

[62]. Reverse Search Warrants, supra note 53.

     [63]. Chelsa Camille Edano, Beware What You Google: Fourth Amendment Constitutionality of Keyword Warrants, 97 Wash. L. Rev. 977, 979–87 (2022).

     [64]. See Jennifer Lynch & Andrew Crocker, Colorado Supreme Court Upholds Keyword Search Warrant, Elec. Frontier Found. (Oct. 16, 2023), https://www.eff.org/deeplinks/2023/10/colorado-supreme-court-upholds-keyword-search-warrant [https://perma.cc/L8GM-ZFZU].

     [65]. See Lynch, supra note 18 (noting that geofence warrants almost always are for Google).

     [66]. Fourth Amend. Ctr., supra note 14, at 3.

     [67]. Zack Whittaker, Google Says Geofence Warrants Make Up One-Quarter of All US Demands, TechCrunch (Aug. 19, 2021), https://techcrunch.com/2021/08/19/google-geofence-warrants/ [https://perma.cc/6FWY-26SX].

     [68]. See Google, supra note 60.

[69]. See United States v. Chatrie, 590 F. Supp. 3d 901, 908–14 (E.D. Va. 2022) (discussing that while users opt into sharing location history data, the choice is not truly voluntary as the user interfaces created by Google make the choice to decline cumbersome), aff’d, 107 F.4th 319 (4th Cir. 2024), aff’d on reh’g en banc, 136 F.4th 100 (4th Cir. 2025), cert. granted in part, 223 L. Ed. 2d 553 (U.S. 2026).

[70]. Id. at 908.

[71]. Id.

[72]. Id.

[73]. Id.

[74]. See United States v. Smith, 110 F.4th 817, 823–24 (5th Cir. 2024).

     [75]. Brief of Amicus Curiae Google LLC in Support of Neither Party Concerning Defendant’s Motion to Suppress Evidence from a “Geofence” General Warrant at 8, United States v. Chatrie, 590 F. Supp. 3d 901 (E.D. Va. 2022) (No. 3:19cr130).

     [76]. Id.

     [77]. SeeChatrie, 590 F. Supp. 3d at 914–16.

[78]. Id. at 915.

[79]. Id. at 915 n.19.

     [80]. See Transcript of Motion to Suppress at 61–63, Chatrie, 590 F. Supp. 3d 901 (No. 3:19CR130).

[81]. See United States v. Smith, 110 F.4th 817, 837 (5th Cir. 2024) (discussing that Google must necessarily search the records of all 592 million people as part of its response to a geofence warrant).

     [82]. Chatrie, 590 F. Supp. 3d at 916.

[83]. Id.

     [84]. Julia Love & Davey Alba, Google User Data Is Police’s Top Shortcut for Solving Crimes, Bloomberg Businessweek (Sep. 28, 2023), https://www.bloomberglaw.com/product/blaw/bloombergterminalnews/bloomberg-terminal-news/S1OGHQT1UM0X?criteria_id=f4f2b3db5ccb9a367eccfae10446bdb9 [https://perma.cc/NFX4-XZ6R].

[85]. Id.

[86]. Id.

     [87]. Id.

[88]. Id.

     [89]. See McGriff, supra note 61.

     [90]. See Lynch, supra note 18 (“These changes would appear to make it much more difficult—if not impossible—for Google to provide mass location data in response to a geofence warrant.”); see also Zack Whittaker, Google Moves to End Geofence Warrants, a Surveillance Problem It Largely Created, TechCrunch (Dec. 16, 2023), https://techcrunch.com/2023/12/16/google-geofence-warrants-law-enforcement-privacy/ [https://perma.cc/8Z5V-RU65] (expressing doubt that geofence warrants will go away as “Google still retains huge banks of historical location data that police can tap into anytime” and “[t]ech companies store vast troves of users’ location data, so they too can be subject to similar legal demands”).

     [91]. For example, in the summer of 2025, Brookline, Massachusetts police investigating a vandalism of a Jewish grocery store asked their FBI counterparts to use a geofence warrant to identify possible suspects. See, e.g., Sam Mintz, FBI Considered Opening Investigation into Jewish Grocery Store Vandalism, Emails Show, Brookline News (Sep. 29, 2025), https://brookline.news/fbi-considered-opening-investigation-into-jewish-grocery-store-vandalism-emails-show/ [https://perma.cc/8BC7-5K3W] (“The FBI considered using a controversial law enforcement technique called a geofence warrant to open a federal investigation into an act of antisemitic vandalism in Brookline, according to public records. . . . Ultimately, the FBI declined to open a case, with an agent citing an inability to use geofencing because of ‘management’ at the agency, according to the records.”). The Brookline Police and the FBI’s actions came to light through a public records request by Nathan Story, who published the emails on his blog. See Nathan Story, Town of Brookline / Butcherie Emails, Nathan Story Blog (Sep. 24, 2025), https://blog.nathanielstory.com/blog/butcherie-emails/ [https://perma.cc/H5AN-5S5W] (hosting public records and describing two requests as “page 282 communication with FBI, Brookline PD Detective advocated ‘attempting to submit a geofence’” and “page 311 FBI responds no to geofence, and ‘We aren’t opening on our end at this time’”).

     [92]. Brief for the United States in Opposition at 18, Chatrie v. United States, 223 L. Ed. 2d 553 (U.S. 2026) (“This Office has been informed that all Location History data was deleted from Google’s Sensorvault database by the end of July 2025. Accordingly, it is the government’s understanding that Google will be unable to respond to geofence warrants going forward, except to the extent that Google retains data responsive to specific geofence warrants or geofence preservation requests received before that date.”).

     [93]. Brief of X Corp. as Amicus Curiae in Support of Petitioner at 1, Chatrie v. United States, 223 L. Ed. 2d 553 (U.S. 2026).

[94].‍ ‍Vidhya Srinivasan, AI, Personalization and the Future of Shopping, Google (Mar. 3, 2025), https://blog.google/products/ads-commerce/ai-personalization-and-the-future-of-shopping/ [https://perma.cc/2P36-3TVM].

[95]. Id.

[96]. Privacy Policy, Google, https://policies.google.com/privacy?hl=en-US#infocollect [https://perma.cc/6HWL-M6XM]; see Lynch & Crocker, supra note 64 (discussing how keyword warrant process works and stating that Google has one billion users with Google accounts).

     [97]. Lynch & Crocker, supra note 64.

[98]. SeeGoogle Trends, Google Trends, https://trends.google.com/trends/explore?geo=US&hl=en [https://perma.cc/362Z-K6AY] (showing “weather” was the most searched term on Google in the United States from March 2024 to March 2025).

[99]. Seeid. (showing “youtube” was the third most searched term on Google in the United States from March 2024 to March 2025).

   [100]. See Teddy Wilson, “Prosecution in Search of a Theory”: Court Documents Raise Questions About Case Against Latice Fisher, Rewire News Grp. (Feb. 21, 2018), https://rewirenewsgroup.com/2018/02/21/prosecution-search-theory-court-documents-raise-questions-case-latice-fisher/ [https://perma.cc/DV78-7LS4] (describing the prosecution of Latice Fisher, a Mississippi woman charged with second-degree murder after experiencing a miscarriage); see also Lynch & Crocker, supra note 64 (“[I]t is virtually impossible to navigate the modern Internet without entering search queries into a search engine like Google’s.”).

   [101]. Ryan Phillips, Infant Death Case Heading Back to Grand Jury, Starkville Daily News (May 9, 2019), https://www.starkvilledailynews.com/infant-death-case-heading-back-to-grand-jury/article_cf99bcb0-71cc-11e9-963a-eb5dc5052c92.html [https://perma.cc/9U8C-ZXVS]. In Latice Fisher’s case, the police did not use a reverse search warrant but instead downloaded data from her cell phone and looked at the Google search results stored on it. This technique, while chilling, is likely less legally problematic than a keyword warrant as here the police had identified a suspect and had probable cause to suspect a crime had been committed under Mississippi law. See Bobby Allyn, Where Abortion is Banned, Someone’s Phone Activity Could Be Used as Criminal Evidence, NPR (June 30, 2022), https://www.npr.org/2022/06/30/1109051837/where-abortion-is-banned-someones-phone-activity-could-be-used-as-criminal-evide [https://perma.cc/7V4B-L9XD] (noting that the case against Latice Fisher was eventually dropped after she was represented by National Advocates for Pregnant Women).

   [102]. See People v. Seymour, 536 P.3d 1260, 1268 (Colo. 2023) (describing how Google has rejected keyword warrants that require it to turn over all data in a non-anonymized fashion as too broad).

[103]. Seeid. (discussing the multiple steps of Google’s response to a keyword search warrant).

[104]. Id.

[105]. Id.

[106]. Seeid. at 1268–69. The Colorado Supreme Court stated that Google turns over “de-identified” lists of responsive searches in step one of the process but also wrote that, in Seymour, the Denver Police sought, and received, IP address information for the searches in step one. It is hard to see how IP address information could be considered “de-identified.” Id.

[107]. Id.

[108]. Id. at 1269.

[109]. Id.

[110]. Id.

   [111]. U.S. Const. amend. IV.

[112]. Id.

[113]. See Illinois v. Gates, 462 U.S. 213, 238 (1983) (setting out a “totality-of-the-circumstances” standard for probable cause determinations in which the judicial officer asks whether “there is a fair probability that contraband or other evidence of a crime will be found in a particular place”).

   [114]. U.S. Const. amend. IV.

   [115]. Cong. Rsch. Serv., Historical Background on Fourth Amendment, Const. Annotated: Libr. of Cong., https://constitution.congress.gov/browse/essay/amdt4-2/ALDE_00013706/ [https://perma.cc/MDX7-ES4R].

[116]. Id.

[117]. Id.

[118]. See Laura K. Donohue, The Original Fourth Amendment, 83 U. Chi. L. Rev. 1181, 1185 (2016) (“The meaning of the rights enshrined in the Constitution provides a critical baseline for understanding the limits of government action—perhaps nowhere more so than in regard to the Fourth Amendment. The Amendment prohibited the government from entering into any home, warehouse, or place of business against the owner’s wishes to search for or to seize persons, papers, and effects, absent a specific warrant. The only exception was when law enforcement or citizens were in active pursuit of a known felon.”); see also Wayne R. LaFave, Search and Seizure: A Treatise on the Fourth Amendment § 1.1(a) (6th ed. 2020).

[119]. See Cong. Rsch. Serv., Adoption of Exclusionary Rule, Const. Annotated: Libr. of Cong., https://constitution.congress.gov/browse/essay/amdt4-7-2/ALDE_00000806/ [https://perma.cc/5MU3-XPWM] (tracing the development of the exclusionary rule).

[120]. See 232 U.S. 383, 392–94 (1914) (holding exclusion of evidence as a remedy when evidence gathered illegally but only applying the rule in federal court).

[121]. See Cong. Rsch. Serv., supra note 46.

[122]. Id.

[123]. See 338 U.S. 25, 27–28 (1949) (“The security of one’s privacy against arbitrary intrusion by the police—which is at the core of the Fourth Amendment—is basic to a free society. It is therefore implicit in ‘the concept of ordered liberty’ and as such enforceable against the States through the Due Process Clause.”).

[124]. See 367 U.S. 643, 655 (1961) (“We hold that all evidence obtained by searches and seizures in violation of the Constitution is, by that same authority, inadmissible in a state court.”).

   [125]. For an oft-criticized search exception, look to the U.S. Supreme Court’s “dog sniff” cases like United States v. Place, 462 U.S. 696, 707 (1983) (holding that a drug-sniffing dog’s sniff of a traveler’s luggage wasn’t a search because it is “sui generis”); see also Ken Lammers, Canine Sniffs: The Search That Isn’t, 1 N.Y.U. J.L. & Liberty 845, 846–49 (2005) (describing history of dog sniff search cases).

   [126]. While seizure analysis is outside the scope of this note, see Drayton for an example of the Court’s logic creating a gap in the protection of the Fourth Amendment. United States v. Drayton, 536 U.S. 194 (2002) (holding that the police boarding a Greyhound bus and asking passengers to consent to searches of luggage was not a seizure under the Fourth Amendment).

[127]. SeeExceptions to Warrant Requirement, Corn. L. Sch.: Legal Info. Inst., https://www.law.cornell.edu/constitution-conan/amendment-4/exceptions-to-warrant-requirement [https://perma.cc/BTX9-E3NJ] (discussing the exigent circumstances, search incident to arrest, vehicle searches, containers in vehicles, and plain view exceptions to the warrant requirement among others).

[128]. SeeExclusionary Rule Overview, Corn. L. Sch.: Legal Info. Inst., https://www.law.cornell.edu/wex/exclusionary_rule [https://perma.cc/3XBF-GM4T] (discussing the good faith, independent source, inevitable discovery, attenuation, impeachment, and qualified immunity exceptions to the exclusionary rule).

   [129]. 468 U.S. 897, 913 (1984).

   [130]. United States v. Chatrie, 590 F. Supp. 3d 901, 925 (E.D. Va. 2022) (relying on good-faith exception not to dismiss charges even though it found geofence warrant to be unconstitutional search), aff’d, 107 F.4th 319 (4th Cir. 2024), aff’d on reh’g en banc, 136 F.4th 100 (4th Cir. 2025), cert. granted in part, 223 L. Ed. 2d 553 (U.S. 2026).

   [131]. United States v. Chatrie, 136 F.4th 100, 101 (4th Cir. 2025) (Diaz, J., concurring) (solely concurring on good-faith-exception grounds), cert. granted in part, 223 L. Ed. 2d 553 (U.S. 2026); id. at 114 (Niemayer, J., concurring) (“In addition, I also concur in the judgment of the court holding that, in any event, law enforcement’s collection of the data from Google was protected because law enforcement relied in good faith on a warrant issued by a detached and neutral judicial officer.”); id. at 115 (King, J., concurring) (“I am pleased to join in the fine concurring opinions of Judge Wilkinson and Judge Richardson. In addition, I agree that the officers acted in good faith, and I therefore also support the affirmance of the district court’s judgment on that basis.”); id. at 115 n.1 (Wynn, J., concurring) (“Although I believe that this case involved a Fourth Amendment search—and that we should say so—I acknowledge that the conditions for application of the good-faith exception to the exclusionary rule are met here.”); id. at 143 (Heytens, J., concurring) (holding good faith rule applies).

   [132]. United States v. Smith, 110 F.4th 817 (5th Cir. 2024) (holding the good-faith exception applies even though the geofence warrant was categorically unconstitutional).

   [133]. Leon, 468 U.S. at 921–22.

[134]. Id. at 916 (“[T]he exclusionary rule is designed to deter police misconduct rather than to punish the errors of judges and magistrates.”).

[135]. Seeid. at 907 (discussing the “social costs” of the exclusionary rule and the court’s prior decisions).

[136]. Seeid. at 928–29 (Brennan, J., dissenting).

[137]. Id. at 923.

   [138]. Id. at 929 (Brennan, J., dissenting) (“[T]oday the Court sanctions the use in the prosecution’s case in chief of illegally obtained evidence against the individual whose rights have been violated—a result that had previously been thought to be foreclosed.”).

   [139]. 110 F.4th 817, 838 (5th Cir. 2024).

   [140]. Id. at 820.

   [141]. Seeid. at 840.

[142]. Seeid.

[143]. Seeid.

   [144]. Initial Brief of Appellant Criminal Appeal at 49, United States v. Smith, 110 F.4th 817 (5th Cir. 2024), 2023 WL 8283119, at *49 (“There is no such thing as relying on a general warrant in good faith. To hold otherwise would invite the kind of systematic error and reckless disregard of constitutional requirements that this Court has cautioned against.” (internal citations and quotation marks omitted)).

   [145]. Justice Marshall, Justice Brennan, Justice Scalia, Justice Sotomayor, and in recent years for digital rights issues, Justice Roberts and Justice Gorsuch, have had significant Fourth Amendment opinions, concurrences, or dissents. See, e.g., United States v. Leon, 468 U.S. 897, 928–960 (1984) (Brennan, J., dissenting) (good-faith exception case); United States v. Jones, 565 U.S. 400, 402–13 (2012) (Scalia majority opinion revitalizing trespass approach to Fourth Amendment); Jones, 565 U.S. at 413–18 (Sotomayor concurrence on REOP grounds); Florida v. Jardines, 569 U.S. 1, 3–12 (2013) (Scalia majority opinion elaborating on property-based approach); Carpenter v. United States, 138 S. Ct. 2206, 2211–24 (2018) (Roberts majority opinion finding warrantless acquisition of cell-site location information to violate REOP); Carpenter, 138 S. Ct. at 2261–72 (Gorsuch, J., dissenting) (advocating for a bailment-based conception of Fourth Amendment analysis).

   [146]. Katz v. United States, 389 U.S. 347, 351 (1967).

[147]. Id.

[148]. Seeid. at 352–53 (“[F]or that Amendment was thought to limit only searches and seizures of tangible property. But ‘[t]he premise that property interests control the right of the government to search and seize has been discredited.’” (quoting Warden v. Hayden, 387 U.S. 294, 304 (1967))).

[149]. Id. at 361 (Harlan, J., concurring).

[150]. See United States v. Jones, 565 U.S. 400, 406 (2012) (“Our later cases have applied the analysis of Justice Harlan’s concurrence in that case, which said that a violation occurs when government officers violate a person’s ‘reasonable expectation of privacy.’”); see alsoState Action Doctrine, Corn. L. Sch.: Legal Info. Inst., https://www.law.cornell.edu/constitution-conan/amendment-14/state-action-doctrine [https://perma.cc/3HKV-DGH5]. Per the State Action Doctrine, the Constitution generally applies only to the government, and there is a vast body of caselaw of when the actions of a private individual are sufficiently close to that of the state to count as state action. Id.

   [151]. Katz, 389 U.S. at 361 (Harlan, J., concurring) (“[T]here is a twofold requirement, first that a person have exhibited an actual (subjective) expectation of privacy and, second, that the expectation be one that society is prepared to recognize as ‘reasonable.’”).

[152]. Id.

   [153]. Seeid.

[154].‍ ‍Harvey A. Scheider, Katz v. United States: The Untold Story,40 McGeorge L. Rev. 13, 13 (2008).

   [155]. Katz, 389 U.S. at 348.

[156]. Id. at 348 n.1.

[157]. Id. at 352.

[158]. Id.

[159]. Seeid. at 351 (announcing the approach taken in the case did not come from the parties’ briefs).

[160]. See Orin S. Kerr, Katz Has Only One Step: The Irrelevance of Subjective Expectations, 82 U. Chi. L. Rev. 113, 118 (2015) (describing empirical study in which researchers analyzed 540 cases that applied the Katz test and found that in “nine out of ten cases, courts applied Katz without even considering whether the defendant had satisfied the subjective test”).

   [161]. Katz, 389 U.S. at 348.

[162]. See, e.g., LaFave, supra note 118, § 2.1(b) (“[I]t can hardly be said that the Court produced clarity where theretofore there had been uncertainty. If anything, the exact opposite has occurred. The pre-Katz rule, though perhaps ‘unjust’ was ‘a workable tool for the reasoning of the courts.’”); see alsoid. (“Katz unquestionably expands the coverage of the Fourth Amendment.”).

[163]. See United States v. Miller, 425 U.S. 435, 440 (1976) (holding that Miller has no possessory interest in bank records and that they are “business records of the banks” because they are related to transactions in which the bank was a party); id. at 442 (“We perceive no legitimate ‘expectation of privacy’ in their contents [checks and deposit slips at bank] . . . . All of the documents obtained, including financial statements and deposit slips, contain only information voluntarily conveyed to the banks and exposed to their employees in the ordinary course of business.”); Smith v. Maryland, 442 U.S. 735, 742 (1979) (“[W]e doubt that people in general entertain any actual expectation of privacy in the numbers they dial. All telephone users realize that they must ‘convey’ phone numbers to the telephone company, since it is through telephone company switching equipment that their calls are completed.”).

[164]. SeeMiller, 425 U.S. at 442.

[165]. Seeid. at 444.

[166]. SeeSmith, 442 U.S. at 747 (Stewart, J., dissenting) (“I think that the numbers dialed from a private telephone—like the conversations that occur during a call—are within the constitutional protection recognized in Katz.”); see also Hanni Fakhoury, Smith v. Maryland Turns 35, But Its Health Is Declining, Elec. Frontier Found. (June 24, 2014), https://www.eff.org/deeplinks/2014/06/smith-v-maryland-turns-35-its-healths-declining [https://perma.cc/W4D8-SZRW] (noting that courts have been rethinking “whether Smith is the absolute rule the government claims it should be”).

   [167]. Smith, 442 U.S. at 735, 744.

[168]. Seeid. at 746 (Stewart, J., dissenting) (observing that a “telephone call simply cannot be made without use of telephone company property”).

[169]. See 565 U.S. 400, 417 (2012) (Sotomayor, J., concurring) (While Jones was decided as a property-approach case, Justice Sotomayor wrote in a concurring opinion: “[I]t may be necessary to reconsider the premise that an individual has no reasonable expectation of privacy in information voluntarily disclosed to third parties . . . . This approach is ill suited to the digital age, in which people reveal a great deal of information about themselves to third parties in the course of carrying out mundane tasks.”).

[170]. See 138 S. Ct. 2206, 2219 (2018) (holding that the third-party doctrine does not apply to CSLI because there is a “world of difference between the limited types of information addressed in Smith and Miller and the exhaustive chronicle of information casually collected by wireless carriers”).

[171]. Id. at 2220.

[172]. Id.

   [173]. Id.

   [174]. Matthew J. Tokson, The Carpenter Test as a Transformation of Fourth Amendment Law, 2023 U. Ill. L. Rev. 507, 523 (2023).

   [175]. U.S. Supreme Court Issues Groundbreaking Victory for Privacy Rights, ACLU Mass. (June 22, 2018), https://www.aclum.org/press-releases/us-supreme-court-issues-groundbreaking-victory-privacy-rights/ [https://perma.cc/M6J4-UHMS]; Andrew Crocker & Jennifer Lynch, Victory! Supreme Court Says Fourth Amendment Applies to Cell Phone Tracking, Elec. Frontier Found. (June 22, 2018), https://www.eff.org/deeplinks/2018/06/victory-supreme-court-says-fourth-amendment-applies-cell-phone-tracking [https://perma.cc/EY32-X2JX].

   [176]. United States v. Chatrie, 107 F.4th 319, 332 (4th Cir. 2024) (“The third-party doctrine squarely governs this case.”), aff’d on reh’g en banc, 136 F.4th 100 (4th Cir. 2025), cert. granted in part, 223 L. Ed. 2d 553 (U.S. 2026).

   [177]. Riley v. California, 573 U.S. 373, 400 (2014).

   [178]. Chatrie, 107 F.4th at 332.

   [179]. United States v. Smith, 110 F.4th 817, 836 (5th Cir. 2024).

   [180]. 565 U.S. 400, 403 (2012).

[181]. Id. at 402.

[182]. Id. at 403.

[183]. Id. at 405.

   [184]. Id. at 404; see Maureen E. Brady, The Lost “Effects” of the Fourth Amendment: Giving Personal Property Due Protection, 125 Yale L.J. 946 (2016) (remarking that before Jones, “effects” had gotten short shrift in Fourth Amendment jurisprudence and that by analyzing effects through the REOP framework, it minimizes the useful realities of property law).

   [185].‍ ‍Jones, 565 U.S. at 404–05.

[186]. Id. at 405–06.

[187]. Id. at 411.

   [188]. Id. at 405.

   [189]. Trespass, Corn. L. Sch.: Legal Info. Inst., https://www.law.cornell.edu/wex/trespass [https://perma.cc/7B79-ETR2].

   [190]. Jones, 565 U.S. at 404–05.

   [191]. 569 U.S. 1, 11–12 (2013).

   [192]. See id. at 7–8; see also Orin S. Kerr, The Two Tests of Search Law: What Is the Jones Test, and What Does That Say About Katz?,103 Wash. U. L. Rev. 309, 327 n.117 (2025) (observing that the U.S. Supreme Court did not use the word “trespass” in Jardines). This test is likely broader than the test in Jones, because under a common law understanding of trespass, the officers likely were not trespassing as they were approaching Jones’s porch—an area anyone would have to pass through to knock on the door. However, scholarly commentary is not clear on whether the Jones test should be understood as trespass or physical intrusion. Professor Orin Kerr argues that Jones, despite its reliance on common law trespass, should be understood as a physical intrusion case, which would more closely align the holdings in Jones and Jardines along physical intrusion lines. Seeid.

[193]. Jardines, 569 U.S. at 4.

[194]. Id. at 6.

[195]. Id.

[196]. See id. at 11.

[197]. See Carpenter v. United States, 138 S. Ct. 2206, 2261 (2018) (Gorsuch, J., dissenting).

[198]. Seeid. at 2268.

[199]. Id. at 2269 (“This approach is quite different from Smith and Miller’s (counter)-intuitive approach to reasonable expectations of privacy; where those cases extinguish Fourth Amendment interests once records are given to a third party, property law may preserve them.”).

[200]. Seeid. at 405 (“Even if Katz may still supply one way to prove a Fourth Amendment interest, it has never been the only way. Neglecting more traditional approaches may mean failing to vindicate the full protections of the Fourth Amendment.”).

   [201]. See United States v. Chatrie, 107 F.4th 319 (4th Cir. 2024), aff’d on reh’g en banc, 136 F.4th 100 (4th Cir. 2025), cert. granted in part, 223 L. Ed. 2d 553 (U.S. 2026).

   [202]. See United States v. Smith, 110 F.4th 817 (5th Cir. 2024).

   [203]. Petition for a Writ of Certiorari, Chatrie v. United States, 223 L. Ed. 2d 553 (U.S. 2026); Petition for a Writ of Certiorari, Smith v. United States, 146 S. Ct. 356 (2025) (No. 24-7237).

   [204]. Petition for a Writ of Certiorari at 1–6, Chatrie v. United States, 223 L. Ed. 2d 553 (U.S. 2026).

   [205]. Chatrie v. United States, 223 L. Ed. 2d 553 (U.S. 2026).

   [206]. Petition for a Writ of Certiorari at 2, Smith v. United States, 146 S. Ct. 356 (2025) (No. 24-7237).

   [207]. Smith v. United States, 146 S. Ct. 356 (2025).

   [208]. See 652 F. Supp. 3d 38 (D.D.C. 2023).

   [209]. See In re Four Applications for Search Warrants Seeking Information Associated with Particular Towers, No. 3:25-CR-38-CWR-ASH, slip op. at 4 (S.D. Miss. Feb. 21, 2025).

   [210]. United States v. Spurlock, 778 F. Supp. 3d 1136, 1138 (D. Nev. 2025).

[211]. Id. at 1138.

   [212]. 536 P.3d 1260, 1276–80 (Colo. 2023) (holding keyword search warrant was sufficiently particularized and supported by probable cause and that good-faith exception applied).

   [213]. See 348 A.3d 133, 156 (Pa. 2025).

   [214]. 913 S.E.2d 700 (Ga. 2025).

   [215]. 714 S.W.3d 614 (Tex. Crim. App. 2025).

   [216]. Wells v. State, No. 25-484 (U.S. Oct. 20, 2025).

[217]. See United States v. Chatrie, 107 F.4th 319, 324 (4th Cir. 2024), aff’d on reh’g en banc, 136 F.4th 100 (4th Cir. 2025), cert. granted in part, 223 L. Ed. 2d 553 (U.S. 2026); United States v. Smith, 110 F.4th 817, 820–21 (5th Cir. 2024).

[218]. See Jeff Williamson, FBI Offering $10,000 Reward for Arrest in Virginia Armed Bank Robbery, WSLS (Virginia) (June 14, 2019), https://www.wsls.com/news/2019/06/14/fbi-offering-10000-reward-for-arrest-in-virginia-armed-bank-robbery/ [https://perma.cc/VK7U-LBNK] (noting the image of person wearing a hi-vis vest while committing robbery).

   [219]. United States v. Chatrie, 590 F. Supp. 3d 901, 905 (E.D. Va. 2022), aff’d, 107 F.4th 319, 324 (4th Cir. 2024), aff’d on reh’g en banc, 136 F.4th 100 (4th Cir. 2025), cert. granted in part, 223 L. Ed. 2d 553 (U.S. 2026).

   [220]. United States v. Smith, No. 3:21-cr-107-SA, 2023 WL 1930747, at *1 (N.D. Miss. 2023).

   [221]. 590 F. Supp. 3d at 906.

   [222]. 2023 WL 1930747, at *1.

[223]. SeeChatrie, 590 F. Supp. 3d at 917 (discussing Detective Hylton’s investigation, including observing the alleged robber using what appeared to be a cell phone in security camera footage); see alsoSmith, 2023 WL 1930747, *1 (discussing Postal Inspector’s affidavit in support of the warrant in which the inspector noted seeing the alleged robber using his cell phone before and after robbery).

   [224]. Chatrie, 590 F. Supp. 3d at 917; Smith, 2023 WL 1930747, at *2.

   [225]. 590 F. Supp. 3d at 917.

   [226]. 2023 WL 1930747, at *2.

   [227]. Chatrie, 590 F. Supp. 3d at 917–22.

   [228]. Smith, 2023 WL 1930747, at *4–5.

[229]. SeeChatrie, 590 F. Supp. 3d at 905; Smith, 2023 WL 1930747, at *5.

   [230]. 590 F. Supp. 3d at 927.

[231]. Id. at 929.

[232]. Id. at 936.

   [233]. See Defendant Okello T. Chatrie’s Motion to Suppress Evidence Obtained from a “Geofence” General Warrant at 14–16, United States v. Chatrie, 590 F. Supp. 3d 901 (E.D. Va. 2022) (No. 3:19cr130) (Part I.B.2.C entitled “The Acquisition of Defendant’s GPS Data from Google Was a Search Under a Property-Based Approach to the Fourth Amendment”). By contrast, the defendant in Smith did not raise a property-based argument in his appellate brief but did discuss it in his reply brief. See Reply Brief of Appellant’s Criminal Appeal at *3–4, United States v. Smith, 110 F.4th 817 (2024) (No. 23-60321), 2024 WL 218086, at *3–4.

   [234]. See 2023 WL 1930747, at *6.

   [235]. Id. at *7–9.

[236]. Id. at *10.

[237]. Id. at *10–12.

[238]. Geofence Warrants and the Grand Canyon of Circuit Splits, supra note 19.

   [239]. United States v. Chatrie, 107 F.4th 319, 322 (4th Cir. 2024), aff’d on reh’g en banc, 136 F.4th 100 (4th Cir. 2025), cert. granted in part, 223 L. Ed. 2d 553 (U.S. 2026).

[240]. Id. at 330.

[241]. Id.

[242]. Id. at 331. Justice Kennedy made this point in his dissent in Carpenter:

Still, the Court maintains, cell-site records are ‘unique’ because they are ‘comprehensive’ in their reach; allow for retrospective collection; are ‘easy, cheap, and efficient compared to traditional investigative tools’; and are not exposed to cell phone service providers in a meaningfully voluntary manner. But many other kinds of business records can be so described. Financial records are of vast scope. Banks and credit card companies keep a comprehensive account of almost every transaction an individual makes on a daily basis.

Carpenter v. United States, 138 S. Ct. 2206, 2232 (2018) (internal citations omitted).

   [243]. Chatrie, 107 F.4th at 331.

[244]. Id. at 337.

   [245]. See id. at 339 (Wynn, J., dissenting).

   [246]. United States v. Chatrie, No. 22-4489, 2024 WL 4648102 (4th Cir. Nov. 1, 2024) (granting rehearing en banc).

   [247]. See United States v. Chatrie, 136 F.4th 100 (4th Cir. 2025), cert. granted in part, 223 L. Ed. 2d 553 (U.S. 2026). The divide between the Fourth and Fifth Circuits is not a “true” circuit split because the en banc Fourth Circuit vacated the panel’s opinion in affirming the district court. See id. at 101; Fourth Cir. Loc. R. 40(e). The en banc court provided no reasoning for its majority decision and split 7-7 if a search occurred. See Chatrie, 136 F.4th at 101, 109, 115, 130, 144. Additionally, the en banc court’s eight concurrences and one dissent reflect a court with deep internal divisions, let alone the differences with the Fifth Circuit. The Supreme Court granting certiorari to hear Chatrie illustrates that this divide is of national importance, even if it is not currently a classic circuit split due to the en banc Fourth Circuit’s vacatur of the panel’s decision.

[248]. Id. at 101.

   [249]. Id.

   [250]. Orin Kerr, The Fourth Circuit’s Geofencing Case Ends Not With a Bang But A Whimper, Reason: Volokh Conspiracy (May 2, 2025), https://reason.com/volokh/2025/05/02/the-fourth-circuits-geofencing-case-ends-not-with-a-bang-but-a-whimper/ [https://perma.cc/RAC9-A6GS].

[251].‍ ‍United States v. Smith, 110 F.4th 817 (5th Cir. 2024).

[252]. Smith, 110 F.4th at 841 (Ho, J., concurring).

[253]. Id. at 838–40.

[254]. Id. at 836.

[255]. Id. at 833 (internal quotations omitted).

[256]. Id. at 834–35.

[257]. Id. at 838.

[258]. Id. at 837–38. But see United States v. Chatrie, 136 F.4th 100, 155 (4th Cir. 2025) (Berner, J., concurring) (noting that the “hundreds of millions” number is a “red herring” because the government does not conduct a search at step 1—it is Google), cert. granted in part, 223 L. Ed. 2d 553 (U.S. 2026). Judge Berner’s analysis relies on a reading of the state action doctrine under which “[t]he government does not search every user in Google’s dataset each time it requests Location History data. A search can occur only when the government accesses the requested information, not when a company begins looking through its internal database.” Id. However, Google only looked through its records because government agents compelled it to do so.

   [259]. United States v. Smith, No. 23-60321 (5th Cir. Jan. 14, 2025).

   [260]. See Orin Kerr, Can Copying Data Be a Fourth Amendment Seizure If Accessing It Is Not a Search?, Reason: Volokh Conspiracy (Oct. 17, 2023), https://reason.com/volokh/2023/10/17/can-copying-data-be-a-fourth-amendment-seizure-if-accessing-it-is-not-a-search/ [https://perma.cc/QN2W-YGK2] (writing that the Colorado Supreme Court addressed reverse warrants in a “very novel way”).

[261].‍ ‍Commonwealth v. Kurtz, 348 A.3d 133, 150 (Pa. 2025).

[262]. See Jones v. State, 913 S.E.2d 700 (Ga. 2025).

   [263]. Wells v. State, 714 S.W.3d 614 (Tex. Crim. App. 2025).

   [264]. People v. Seymour, 536 P.3d 1260, 1267–68 (Colo. 2023).

[265]. Id.

[266]. Id. at 1268.

[267]. Id.

[268]. Id.

[269]. Id. at 1268–69.

[270]. Id. at 1269.

[271]. Id.

[272]. Id.

[273]. Id.

[274]. Id.

[275]. Id.

   [276]. Id.

   [277]. Id. at 1267.

[278]. Id.

   [279]. Id. The founders intended for the Fourth Amendment to protect the First Amendment, and nowhere is this more acutely felt than in a person’s “papers.” See Michael W. Price, Rethinking Privacy: Fourth Amendment “Papers” and the Third-Party Doctrine, 8 J. Nat’l Sec. L. & Pol’y 247, 257 (2016) (“By design, therefore, a paramount purpose of the Fourth Amendment was to serve as a guardian of individual liberty and free expression.”).

[280]. Seymour, 536 P.3d at 1268.

[281]. Id.

[282]. See id. at 1273.

[283]. Seeid. at 1270–73 (discussing Seymour’s reasonable expectation of privacy in his Google search records).

[284]. Seeid. at 1273–74 (analyzing Seymour’s property interest in his Google search results).

   [285]. Compare id. at 1270–73 (multipage discussion of REOP), with id. at 1273–74 (single-page discussion of property).

   [286]. Id. at 1270–71 (“The first (or subjective) prong is an important, but not dispositive, element. Here, Seymour evinced his subjective expectation of privacy in his Google search history by using a search engine with strict privacy protections.” (internal citations and quotation marks omitted)); id. at 1272 (“Accordingly, we conclude that Seymour has a reasonable expectation of privacy in his Google search history under article II, section 7 of the Colorado Constitution.”).

[287]. Id. at 1271–72.

[288]. Id.

[289]. Id. at 1272–73.

[290]. Id. at 1271.

   [291]. Id. at 1272; see also Kerr, supra note 260 (“First, it held that although it was likely not a Fourth Amendment ‘search’ to collect that information—the user is sending the data to Google, after all, so Google receiving and storing it is covered by the third-party doctrine—it is a search under the broader Colorado Constitution.”).

   [292]. SeeSeymour, 536 P.3d at 1281 n.1 (Berkenkotter, J., concurring ) (“I am also concerned that the majority’s reasoning throughout its opinion rests on a somewhat fuzzy understanding of the technology at play in this case and others that raise questions about the boundaries of the Fourth Amendment.”).

   [293]. Id. at 1272.

[294]. Id.

   [295]. See Kerr, supra note 260 (discussing the “novel” approach the Colorado Supreme Court took in Seymour and suggesting the court found a broader protection from seizures than searches); Seymour, 536 P.3d at 1273.

[296]. SeeSeymour, 536 P.3d at 1273 (citing Ruckelshaus v. Monsanto Co., 467 U.S. 986, 1003 (1984), for the proposition that property rights extend to nontangible property).

[297]. Id.

   [298].‍ ‍Id. (citing Rakas v. Illinois, 439 U.S. 128, 143 n.12 (1978)).

[299]. Id.

[300]. Id.

[301]. Id. at 1273–74.

   [302]. Kerr, supra note 260.

   [303]. SeeSeymour, 536 P.3d at 1273–74.

   [304]. Id. at 1272–73 (holding that the police’s actions were not a search under the meaning of the Fourth Amendment).

[305]. See id. at 1282 (Marquez, J., dissenting).

[306]. See Kerr, supra note 260 (“The problem is that the process of revealing information almost always requires first copying it. If you say that it’s a seizure to copy information, even when revealing it is not a search, then you end up requiring a warrant for act of obtaining the data even though the obtaining doesn’t require a warrant.”).

[307]. See id.

[308]. See id.

   [309]. Seymour, 536 P.3d at 1271, 1273 n.8 (“Here, although Google employees, rather than law enforcement officers, initially searched the database and copied the relevant data, they did so only in response to the warrant and because law enforcement couldn’t access the database directly to conduct the search. Thus, the constitutional protections still apply.”).

   [310]. Commonwealth v. Kurtz, 348 A.3d 133, 156 (Pa. 2025).

   [311]. Id. at 166 (Donohue, J., dissenting) (“Kurtz does not assert that this is a search based on a property-based interest in his Google search queries. Instead, he is focused on a test developed in the 1960s in Katz v. United States.”).

   [312]. Id. at 138–39.

   [313]. Id.

   [314]. Id.

   [315]. Id. at 150.

   [316]. Id.

   [317]. Id.

   [318]. Id. at 145–52.

   [319]. Id.

   [320]. Id. at 149.

   [321]. Id.

   [322]. Id.

   [323]. Jones v. State, 913 S.E.2d 700, 706 n.1 (Ga. 2025) (“In granting Jones’s application for an interlocutory appeal, we also asked the parties to address the threshold question whether the government conducts a search under the Fourth Amendment when it accesses a person’s cell-phone-GPS-based location history. . . . Having reviewed the record, however, it is now apparent that both the trial court and the parties, including the State, took as a given that a search under the Fourth Amendment was carried out here.”).

   [324]. Id. at 703.

   [325]. See United States v. Smith, 110 F.4th 817, 837 (5th Cir. 2024) (“Geofence warrants present the exact sort of ‘general, exploratory rummaging’ that the Fourth Amendment was designed to prevent.”).

   [326]. See United States v. Chatrie, 136 F.4th 100, 101(4th Cir. 2025) (affirming panel’s holding that there was no search), cert. granted in part, 223 L. Ed. 2d 553 (U.S. 2026).

[327]. See Jones, 913 S.E.2d at 708.

[328]. Id.

[329]. Id.

   [330]. See Love & Alba, supra note 84.

   [331]. Like the Georgia Supreme Court, the Texas Court of Criminal Appeals did not discuss users’ property interests in their data. See Wells v. State, 714 S.W.3d 614, 620 (Tex. Crim. App. 2025).

   [332]. Id. at 615–16, 620.

[333]. Id. at 624.

[334]. Id. at 624–25.

[335]. See id. at 636 (Newell, J., concurring in part and dissenting in part).

[336]. SeeGeofence Warrants and The Fourth Amendment, supra note 18, at 2521 (“If law enforcement needed to establish only probable cause to search a private company’s location history records, probable cause would always be satisfied with the same choice statistics and cases about cell phone usage.”).

   [337]. Wells, 714 S.W.3d 614 at 626.

   [338]. See, e.g., Pollack & Tokson, supra note 38, at 711 (arguing that “[p]roperty law is best suited for a background role in Fourth Amendment cases” because it is complex, vulnerable to government manipulation, difficult to apply to new technologies, and privileges those who already own property).

   [339]. See, e.g., United States v. Chatrie, 136 F.4th 100, 100 (4th Cir. 2025) (en banc) (vacating a panel decision holding that geofence warrants are not a search and affirming a district court decision that geofence warrants require probable cause, but that the good faith exception applies), cert. granted in part, 223 L. Ed. 2d 553 (U.S. 2026); United States v. Smith, 110 F.4th 817, 837–38 (5th Cir. 2024) (holding geofence warrants to be an unconstitutional general warrant); Wells v. State, 714 S.W.3d 614, 616 (Tex. Crim. App. 2025) (holding geofence warrants to be constitutional); Jones v. State, 913 S.E.2d 700, 703 (Ga. 2025) (holding that a geofence warrant satisfied probable cause and particularity requirements); Commonwealth v. Kurtz, 348 A.3d 133, 155–56 (Pa. 2025) (holding no REOP in internet search history under Federal or Pennsylvania constitutions); People v. Seymour, 536 P.3d 1260, 1273–74 (Colo. 2023) (holding that a keyword search warrant was a search under the Colorado constitution and a seizure under Federal and Colorado constitutions).

   [340]. See Price, supra note 279, at 248 (arguing for a new Fourth Amendment framework based on “papers” that does not displace existing protections).

   [341]. See Sam Kamin, Katz and Dobbs: Imagining the Fourth Amendment Without a Right to Privacy, 101 Tex. L. Rev. Online 80, 81 (2022) (“As a result, many of the bedrocks of the last half century of Fourth Amendment jurisprudence may now be as vulnerable as Roe, Casey, and the Substantive Due Process decisions directly targeted by Justice Thomas.”).

   [342]. Carpenter v. United States, 138 S. Ct. 2206, 2261–72 (2018) (Gorsuch, J., dissenting).

   [343]. See, e.g., D’Onfro & Epps, supra note 36, at 972–80 (discussing the Fourth Amendment and bailment); Shane Gallant, The Old Bailment Doctrine: The Answer to Fourth Amendment Jurisprudence in the Digital Age, 25 Roger Williams U. L. Rev. 116, 144 (2020) (discussing “intermediary” and “adhesion” bailments in context of the Fourth Amendment); Block & Paul, supra note 29, at 197 (discussing user ownership of search history data in context of keyword search warrants); Kamin, supra note 341, at 97–100 (building on Justice Gorsuch’s bailment idea); Brady, supra note 34, at 1040–43 (offering qualified criticisms of “general law” approach to Fourth Amendment and discussing difficulties of applying bailment doctrine to digital data); Pollack & Tokson, supra note 38, at 708–11 (offering criticisms of property as Fourth Amendment doctrine and discussing difficulties of applying bailment doctrine); Donohue, supra note 38, at 389–400 (in-depth discussion of bailment as it applies to digital data).

   [344]. See, e.g., Baude & Stern, supra note 35, at 1825; D’Onfro & Epps, supra note 36, at 920–26; Brady, supra note 34, at 1010–11, 1016–18.

   [345]. See, e.g., D’Onfro & Epps, supra note 36, at 927. But see Brady, supra note 34, at 1016–42; Pollack & Tokson, supra note 38, at 720–21.

   [346]. See, e.g., People v. Seymour, 536 P.3d 1260 (Colo. 2023); Dalia Wrocherinsky, Finding Rights in the Fine Print: How Terms of Services Agreements Can Turn Consumer Search History into Digital Property, 14 Am. U. Bus. L. Rev. 501, 503 (2024). But see Orin S. Kerr, Terms of Service and Fourth Amendment Rights, 172 U. Pa. L. Rev. 287, 290 (2023); Block & Paul, supra note 29, at 200 n.83.

   [347]. See, e.g., Price, supra note 279, at 270; Donohue, supra note 38, at 391.

   [348]. Seymour, 536 P.3d at 1260.

   [349]. See Price, supra note 279, at 270 (arguing that “entire categories of data, such as communications data and cloud data” should be considered “papers” under the Fourth Amendment for the expressive and associational content contained within them and acknowledging that “the history and purpose of the Fourth Amendment . . . dictate[s] a duty to safeguard expressive and associational materials from unreasonable government intrusion”); Carpenter v. United States, 138 S. Ct. 2206, 2272 (2018) (Gorsuch, J., dissenting) (“It seems to me entirely possible a person’s cell-site data could qualify as his papers or effects under existing law.”).

   [350]. SeeCarpenter, 138 S. Ct. at 2211–23 (majority opinion).

   [351]. See id. at 2261–72 (Gorsuch, J., dissenting).

   [352]. See, e.g., Michael J. O’Connor, Digital Bailments, 22 U. Pa. J. Const. L. 1271, 1273–75 (2020) (discussing shortcomings of REOP and benefits of property approach); D’Onfro & Epps, supra note 36, at 925 (describing scholars’ turn to a “positive law” model of Fourth Amendment in wake of Gorsuch’s opinion in Carpenter); Kamin, supra note 341, at 81.

   [353]. See Carpenter, 138 S. Ct. at 2268 (Gorsuch, J., dissenting).

   [354]. Id.

   [355]. See, e.g., Orin S. Kerr, Katz as Originalism, 71 Duke L.J. 1047, 1090 (2022) (“After confidently announcing that a real originalist and textualist test had existed, Gorsuch can’t come up with a description of it.”); Pollack & Tokson, supra note 38, at 723 n.112 (describing Justice Gorsuch’s opinion in Carpenter as “briefly critiquing Katz on originalist grounds but declining to offer historical support for a property-based approach beyond cursory allusions to the Fourth Amendment’s text, and then noting the further work needed to answer the key question of ‘what kind of legal interest is sufficient to make something yours’”).

   [356]. Carpenter, 138 S. Ct. at 2268 (Gorsuch, J., dissenting).

   [357]. Id.

   [358]. Id. at 2269.

   [359]. Id. at 2270.

   [360]. Id.

   [361]. Id. at 2271; see also Kerr, supra note 355, at 1089–94 (arguing that Justice Gorsuch’s approach collapses back into the Katz REOP framework).

[362]. SeeCarpenter, 138 S. Ct. at 2269 (Gorsuch, J., dissenting).

[363]. Bailment, Black’s Law Dictionary (12th ed. 2024).

[364]. SeeCarpenter, 138 S. Ct. at 2268–69 (Gorsuch, J., dissenting).

[365]. Seeid. at 2270–71.

   [366]. Ex parte Jackson, 96 U.S. 727 (1877).

[367]. SeeCarpenter, 138 S. Ct. at 2269–71 (Gorsuch, J., dissenting).

   [368]. Ex parteJackson, 96 U.S. at 733 (“Letters and sealed packages of this kind in the mail are as fully guarded from examination and inspection, except as to their outward form and weight, as if they were retained by the parties forwarding them in their own domiciles. The constitutional guaranty of the right of the people to be secure in their papers against unreasonable searches and seizures extends to their papers, thus closed against inspection, wherever they may be.”).

   [369]. United States v. Warshak, 631 F.3d 266, 285 (6th Cir. 2010) (“While a letter is in the mail, the police may not intercept it and examine its contents unless they first obtain a warrant based on probable cause. This is true despite the fact that sealed letters are handed over to perhaps dozens of mail carriers, any one of whom could tear open the thin paper envelopes that separate the private words from the world outside. Put another way, trusting a letter to an intermediary does not necessarily defeat a reasonable expectation that the letter will remain private.” (internal citations omitted)).

   [370]. Chris Conley, Tech. & C.L. Project, ACLU of N. Cal., Metadata: Piecing Together a Privacy Solution 1 (2014) (“Generally speaking, the government cannot record or obtain the contents of your communications without at least a search warrant. But ‘metadata,’ information other than communications content, is often treated differently under the law.”).

   [371]. See Price, supra note 279, at 296 (“It took the rise of the Pony Express, founded in 1860, for the Court to establish a Fourth Amendment privacy right in ‘papers’ that are not physically present in the home [in Ex parteJackson].”).

   [372]. See, e.g., D’Onfro, supra note 8, at 150; Gallant, supra note 343, at 119–20; Donohue, supra note 38, at 353; D’Onfro & Epps, supra note 36, at 972 n.335 (noting that Justice Gorsuch’s dissent has inspired many student notes).

   [373]. See Brady, supra note 34, at 1041.

   [374]. D’Onfro & Epps, supra note 36, at 972 (collecting examples of tangible bailments).

   [375]. Id. at 977 (“Although intangible, digital files should be understood as goods that can be delivered to and accepted by a bailee for safekeeping.”); Donohue, supra note 38, at 394 (stating that historically a “bailee had possession, while the bailor retained residual ownership rights” and that this approach maps onto digital data).

   [376]. See Denise R. Johnson, Reflections on the Bundle of Rights, 32 Vt. L. Rev. 247, 247 (2007).

   [377]. See D’Onfro & Epps, supra note 36, at 973 (“The bailee being in possession of the goods yet potentially still subject to the bailor’s right to exclude has important implications for analyzing Fourth Amendment searches under a general-law approach.”).

   [378]. People v. Seymour, 536 P.3d 1260, 1273–74 (Colo. 2023).

   [379]. Donohue, supra note 38, at 394.

   [380]. Id. at 397.

   [381]. Id. Donohue also argues that CSLI fits into “work on a thing” and “keeping of a thing” bailment categories as telecommunications providers use customer data to provide services and maintain their network. The third kind of bailment (“keeping of a thing”) is closer to Ex parte Jackson. See Donohue, supra note 38, at 398. An additional theory of bailment for data targeted by reverse warrants is an “adhesion bailment.” See Gallant, supra note 343, at 144–45. This type of bailment, of Gallant’s own theory, is grounded in a realist conception of data collection: “By merely living in an IoT [internet of things] world, one is continuously providing data to a third-party; it is difficult to understand how anyone can honestly state that such information is voluntarily conveyed. . . . Accordingly, the ‘adhesion bailment doctrine’ is triggered when a third-party company collects data on individuals through means of the user’s day-to-day activities—essentially, the user will automatically have Fourth Amendment protection in the information third-party companies collect through the user’s daily routine.” Id.

   [382]. Brief of Amicus Curiae Google LLC in Support of Neither Party Concerning Defendant’s Motion to Suppress Evidence from a “Geofence” General Warrant at 3, United States v. Chatrie, 590 F. Supp. 3d 901 (E.D. Va. 2022) (No. 3:19cr130) (“Moreover, LH information can often reveal a user’s location and movements with a much higher degree of precision than CSLI and other types of data.”).

   [383]. Seymour, 536 P.3d at 1273.

   [384]. Pollack & Tokson, supra note 38, at 753.

   [385]. Carpenter v. United States, 138 S. Ct. 2206, 2270 (2018) (Gorsuch, J., dissenting).

   [386]. Baude & Stern, supra note 35, at 1825.

   [387]. Id. at 1823.

   [388]. Id.

   [389]. Carpenter, 138 S. Ct. at 2270.

   [390]. See D’Onfro & Epps, supra note 36, at 914 (“Similarly, under Baude and Stern’s model, government only ‘searches’ or ‘seizes’ when it violates the positive law.”).

   [391]. See id. at 917.

   [392]. See id. at 913.

   [393]. Id. at 910, 913.

   [394]. Id. at 931.

   [395]. See Brady, supra note 34, at 1018.

   [396]. See id. at 1019.

   [397]. See id. at 1021 (“Most basically, because the general law in their Article can be derived not only from prevailing case law but also from minority approaches and a range of statutes and social norms, it would seem to be radically open-ended to an early American audience (and likely also to the authors on whom D’Onfro and Epps rely).”).

   [398]. Id. at 1035.

   [399]. People v. Seymour, 536 P.3d 1260, 1273–74 (Colo. 2023) (analyzing Google’s ToS in property-based approach to Fourth Amendment); Byrd v. United States, 138 S. Ct. 1518, 1528–30 (2018) (analyzing rental car contract in context of Fourth Amendment); Commonwealth v. Kurtz, 348 A.3d 133, 150 (Pa. 2025).

   [400]. See, e.g., Kerr, supra note 346, at 503; Wrocherinsky, supra note 346, at 290; Block & Paul, supra note 29, at 200 n.83.

   [401]. 536 P.3d at 1273.

   [402]. Id.

   [403]. 348 A.3d at 150.

   [404]. Orin S. Kerr, Are There Fourth Amendment Rights in Google Search Terms?, Reason: Volokh Conspiracy (Dec. 16, 2025), https://reason.com/volokh/2025/12/16/are-there-fourth-amendment-rights-in-google-search-terms/ [https://perma.cc/927H-TSYF] (“I didn’t think the seizure ruling was persuasive, as I blogged about here, but Wecht opinion here and the Colorad[o] opinion are a fascinating contrast. One uses Google’s TOS to say that a warrant is required; the other uses TOS to say that no warrant is required.”).

   [405]. Kerr, supra note 346, at 290–91.

   [406]. See id. at 290 (“From a practical perspective, this Article has an important doctrinal payoff: it secures Fourth Amendment rights online against the threat of nullification by Terms of Service.”).

   [407]. See id. at 294–99 (discussing cases where courts found that “[t]erms [of service] often reduce or entirely eliminate otherwise existing constitutional protections”).

   [408]. Id. at 294.

   [409]. Seeid. at 328 (“Terms of Service are private contracts, not agreements with the government. They have little or no impact on Fourth Amendment rights.”). But seeid. at 318 (“Terms might, at the margins, clarify certain relationships relevant to Fourth Amendment rights. But it is the actual relationships, not the Terms of Service, that matter.”).

   [410]. People v. Seymour, 536 P.3d 1260, 1273–74 (Colo. 2023) (discussing Google ToS to find that user had possessory interest in internet search history).

   [411]. Kerr, supra note 346, at 287 (“The language in Terms of Service typically gives Internet providers broad rights to address potential account misuse. But do these Terms alter Fourth Amendment rights, either diminishing or even eliminating constitutional rights in Internet accounts?”).

   [412]. Kerr, supra note 404.

   [413]. Wrocherinsky, supra note 346, at 515.

   [414]. Id. at 516.

   [415]. Id. at 516–17.

   [416]. See People v. Seymour, 536 P.3d 1260, 1271 (Colo. 2023).

   [417]. Block & Paul, supra note 29, at 200 (“[U]sers ‘own’ their search data within the meaning of the Fourth Amendment because current positive law––the SCA––grants them a right to exclude others from it.”).

   [418]. See id. at 200 n.83.

   [419]. The court made plain in a footnote that it explicitly relied upon the ToS to find a property interest, but limited its holding to the specific Google ToS at issue. See Seymour, 536 P.3d at 1273 n.7 (“When addressing Seymour’s possessory interest, our opinion is limited to Google accounts and derived from an examination of company-specific policies, terms, and agreements. The discussion on this point should not be taken to address other online search entities, whose user terms may differ from those at issue here.”).

   [420]. However, in Seymour’s case, the Colorado Supreme Court ruled that the good-faith exception applied. See id. at 1281.

   [421]. See Kerr, supra note 404 (describing the split between the Colorado Supreme Court and the Pennsylvania Supreme Court as “a fascinating contrast” and noting that “it’s interesting to see courts rely on [ToS] especially to reach opposite results”).

   [422]. See Kamin, supra note 341, at 82 (“Given what we know about the current Court––particularly its affection for textual and originalist readings of the Constitution and its manifest skepticism of privacy as a constitutional metric––Justice Gorsuch’s Carpenter dissent will become required reading in the event that Katz soon joins Roe and Casey on the ash heap of history.”).

   [423]. See Carpenter v. United States, 138 S. Ct. 2206, 2236 (2018) (Thomas, J., dissenting) (“The more fundamental problem with the Court’s opinion, however is its use of the ‘reasonable expectation of privacy’ test, which was first articulated by Justice Harlan in Katz v. United States. The Katz test has no basis in the text or history of the Fourth Amendment. And it invites courts to make judgments about policy, not law. Until we confront the problems with this test, Katz will continue to distort Fourth Amendment jurisprudence.” (internal citations omitted)); see id. at 2265 (Gorsuch, J., dissenting) (“Even taken on its own terms, Katz has never been sufficiently justified.”).

   [424]. Additionally, Professor Maureen Brady highlights equitable concerns that might animate a ToS theory. Brady, supra note 34, at 1042 (“There are forceful arguments that, in light of the power dynamics between users and major digital file storage companies, courts should modify the law of bailments to protect consumers from the ordinary harshness of contract law; perhaps such disclaimers [disclaiming liability or bailment] should be deemed against public policy.”).

   [425]. Pollack & Tokson, supra note 38, at 722 (describing a set of approaches to the Fourth Amendment as “property-inspired” because “they share both an appreciation of property law as a traditional source of Fourth Amendment protection and a willingness to transform or set aside existing property concepts where necessary” and highlighting Justice Gorsuch’s proposal in Carpenter).

   [426]. See U.S. Const. amend. IV (“The right of the people to be secure in their persons, houses, papers, and effects, shall not be violated.”).

   [427]. See Price, supra note 279, at 249 (“Section III proposes a new, supplemental Fourth Amendment analysis centered on the privacy of one’s ‘papers,’ which enjoy equal billing with ‘persons,’ ‘houses,’ and ‘effects’ in the text, if not in practice.”).

   [428]. SeeCarpenter, 138 S. Ct. at 2272 (Gorsuch, J., dissenting).

   [429]. Id. at 2230 (Kennedy, J., dissenting).

   [430]. See id. at 2222 (internal quotations omitted).

   [431]. See Price, supra note 279, at 270–71.

   [432]. See id. at 271.

   [433]. Id. at 271–72 (“In order to illustrate how Fourth Amendment ‘papers’ might expand to include data held by a third party, consider the evolution of the Supreme Court’s jurisprudence on ‘houses.’ A literal reading of the term could limit the scope of the Fourth Amendment quite significantly. Yet the Court has consistently extended constitutional protection far beyond the four walls of a private residence to include garages, boarding houses, rented houses, hotel rooms, park cabins, factories, private offices, and mobile homes.”).

   [434]. Id. at 274.

   [435]. Id.

   [436]. Carpenter v. United States, 138 S. Ct. 2206, 2227 (2018) (Kennedy, J., dissenting) (“[T]he Fourth Amendment’s protections must remain tethered to the text of that Amendment, which again, protects only a person’s own ‘persons, houses, papers, and effects.’”).

   [437]. Donohue, supra note 38, at 353 (“In ascertaining who owns digital documents and records, the Court can employ a but for analysis, asking whether the material would exist but for the right-holder’s actions.”).

   [438]. For an example on the divide on data ownership, both Professor Donohue and Professors Tokson and Pollack cite to Professor Pamela Samuelson’s article Privacy as Intellectual Property for opposing propositions regarding data ownership. Id. at 391 n.213 (stating that the Professor Samuelson article argues that people think about personal data as property); Pollack & Tokson, supra note 38, at 736 (“As Professor Pamela Samuelson has put it, ‘[T]he traditional view in American Law has been that information as such cannot be owned by any person.’”).

   [439]. Donohue, supra note 38, at 353.

   [440]. Id.

   [441]. Id. at 391.

   [442]. Id. at 409.

   [443]. See id. at 353.

   [444]. See Price, supra note 279, at 270–71.

   [445]. Brief of Amicus Curiae Google LLC in Support of Neither Party Concerning Defendant’s Motion to Suppress Evidence from a “Geofence” General Warrant at 18, United States v. Chatrie, 590 F. Supp. 3d 901 (E.D. Va. 2022) (No. 3:19cr130).

   [446]. See People v. Seymour, 536 P.3d 1260, 1274 (Colo. 2023).

   [447]. Id.

   [448]. Seesupra Parts IV.A.1, IV.A.2.

   [449]. Seesupra Part IV.A.3.

   [450]. Seesupra Part IV.A.4.

   [451]. Seymour, 536 P.3d at 1273–74.

   [452]. See id.

   [453]. See Donohue, supra note 38, at 353–54 (“CSLI closely mirrors a bailment in location rei, in which a considerable amount of control is provided to the possessor, without altering the right-holder’s power—in this case, over his or her location information.”).

   [454]. See Seymour, 536 P.3d at 1274; see also Wrocherinsky, supra note 346, at 501 (arguing that courts should rely on ToS agreements to bolster Fourth Amendment protections in context of reverse keyword search warrants).

   [455]. Seesupra Part IV.A.2.

   [456]. See Katz v. United States, 389 U.S. 347, 361 (1967) (Harlan, J., concurring).

   [457]. Carpenter v. United States, 138 S. Ct. 2206, 2268 (2018).

   [458]. United States v. Chatrie, 590 F. Supp. 3d 901, 914–15 (E.D. Va. 2022) (internal citations and quotation marks omitted), 107 F.4th 319 (4th Cir. 2024), aff’d on reh’g en banc, 136 F.4th 100 (4th Cir. 2025), cert. granted in part, No. 25-112, 2026 WL 120676 (U.S. Jan. 16, 2026).

   [459]. Id. at 915 (internal citations and quotation marks omitted).

   [460]. United States v. Smith, 110 F.4th 817, 837 (5th Cir. 2024).

   [461]. Id.

   [462]. Fan, supra note 12, at 934 (“The rise of new digital search strategies exemplified by keyword and geofence warrants is both tempting and terrifying in its power, and it is spurring litigation with important implications for the future of constitutional criminal procedure.”).

[463]. In re Four Applications for Search Warrants Seeking Information Associated with Particular Towers, No. 3:25-CR-38-CWR-ASH, slip op. at 14 (S.D. Miss. Feb. 21, 2025).

[464]. See Srinivasan, supra note 94.

[465]. See United States v. Smith, 110 F.4th 817, 823–24 (5th Cir. 2024).

   [466]. See United States v. Chatrie, 136 F.4th 100, 100 (4th Cir. 2025) (fractured, single-sentence en banc affirmance with eight concurrences and one dissent), cert. granted in part, 223 L. Ed. 2d 553 (U.S. 2026).

[467]. SeeSmith, 110 F.4th at 840; Chatrie, 136 F.4th at 100.

   [468]. Jones v. State, 913 S.E.2d 700, 703–04 (Ga. 2025).

   [469]. Wells v. State, 714 S.W.3d 614, 616 (Tex. Crim. App. 2025).

   [470]. People v. Seymour, 536 P.3d 1260, 1268 (Colo. 2023).

   [471]. Smith, 110 F.4th at 840.

   [472]. Wells, 714 S.W.3d at 616.

   [473]. See Katz v. United States, 389 U.S. 347, 360 (1967) (Harlan, J., concurring).

   [474]. Seymour, 536 P.3d at 1268.