WELLS v. STATE (2025)

This case involves a question about the constitutionality of a “geofence” warrant.¹ We conclude that use of the geofence warrant in this case to obtain location history data did not violate the Fourth Amendment of the United States Constitution.² Accordingly, the judgment of the court of appeals is affirmed.

I.The Offense

Jimmy Giddings was a drug dealer. He lived with his girlfriend, Nikita Dickerson, at a house at 4923 Veterans Drive in Dallas, across the street from Carver Heights Baptist Church. Dickerson and Giddings had a routine. When he returned home in the early morning hours, she would unlock the gate at their front door and greet him in the driveway. She would carry a .40 caliber Glock pistol because, while they lived in a nice house, she felt the neighborhood was unsafe.

At around 3 a.m. on the morning of the offense, June 24, 2018, Dickerson exited the gate outside the front door, as captured on the home's front-door security camera, pursuant to her and Giddings’ routine. Security cameras from the church across the street recorded four men who had been loitering in the parking lot on the far side of the church from Veteran's Drive “for some hours” before the offense. When Giddings arrived home, the four men, wearing masks over their lower faces, rushed across the street toward Giddings and Dickerson brandishing pistols and a rifle.

In the melee that followed, Dickerson sustained five non-life-threatening gunshot wounds. She also dropped her pistol, and it was retrieved by one of the masked men. At the same time, Giddings fled into the house. Two of the assailants rushed in after him, and a third assailant marched the wounded Dickerson into the house at gunpoint. The fourth man, who turned out to be Appellant, quickly followed them.

All the men except for Appellant had visibly distinctive tattoos. Once inside, during the robbery, one of the assailants—the record does not definitively establish which one—shot Giddings in the neck, severing his spine. As a result of this gunshot wound, Giddings died.

Afterwards, the assailants fled back across the street to their vehicle in the church parking lot and drove off. As described by the court of appeals:

Based on the security camera recording timestamp and footage showing that the men were in the area of the church immediately before and after the offense, [police] obtained a warrant to search Google's records for information on devices located within a rectangular geofence encompassing [Giddings and Dickerson's] house and the portion of the church directly across the street between 2:45 a.m. and 3:10 a.m. on June 24. Ultimately, a cellular phone associated with [A]ppellant was identified as being at the scene. Through [A]ppellant's phone records and a search of social media, police were able to identify Milton Prentice, Brian Groom, and Kiante Watkins as the other three men involved in the offense.

Wells, 675 S.W.3d at 819. Watkins testified as an accomplice witness against Appellant at trial, describing the robbery in some detail.

Appellant was charged with and convicted of the capital murder—during the course of a robbery—of Jimmy Giddings. Tex. Penal Code § 19.03(a)(2). Because the State did not seek the death penalty, Appellant received an automatic sentence of life without parole, without the necessity of a punishment hearing. Tex. Penal Code § 12.31(a)(2).

II. Background

A. The Geofence Warrant

The warrant at issue in this case was directed to “Google LLC[.]” It ordered Google to turn over to the police “GPS, WiFi or Bluetooth sourced location history data” corresponding to “Initial Search Parameters” generated from devices that Google's electronic records showed to have been within certain, particularly circumscribed time and location specifications.³ The warrant required disclosure in three steps.

In Step One, the warrant commanded, “[f]or each location point within the ‘Initial Search Parameters’, Google shall produce anonymized information specifying the corresponding unique device ID, timestamp, coordinates, display radius, and data source, if available (the ‘Anonymized List’)[.]” Police were then to “analyze this location data to identify users who may have witnessed or participated” in the capital offense and “seek any additional information regarding these devices from Google.”

In Step Two, the warrant provided that, “[f]or those accounts identified as relevant to the ongoing investigation through analysis of” the Anonymized List, Google “shall provide additional location history outside of the predefined area for those relevant accounts to determine path of travel.” It then specified that, “[t]his additional location history shall not exceed 60 minutes plus or minus the first and last timestamp associated with the account in the initial dataset.” This step was intended to aid the police in ruling out any devices flagged by the Anonymized List so that the identity of obvious non-witnesses and non-participants would not be revealed.

Finally, in Step Three, the warrant ordered that, “[f]or those accounts identified as relevant to the ongoing investigation through an analysis of provided records, and upon demand,” Google “shall provide the subscriber's information for those relevant accounts to include, subscriber's name, email address, services subscribed to, last 6 months of IP history, SMS account number, and registration IP.”⁴ In other words, only in the last step was sufficient information revealed permitting law enforcement to identify witnesses to, or participants in, the capital offense under investigation. At no point during this three-step process were police required to return to the magistrate for incremental authorization.

B. The Warrant Affidavit

The warrant affidavit started out by providing the “Initial Search Parameters”: a “[g]eographical area identified as a polygon defined by” four “latitude/longitude coordinates and connected by straight lines[,]” as specified.⁵ The affidavit sought “GPS, WiFi or Bluetooth sourced location history data from devices that reported a location” within the described polygon at a window of time within which the capital murder occurred, namely: “June 24, 2018 0245 hrs (2:45 a.m.) to June 24, 2018 0310 hrs (3:10 a.m.) Central Time Zone[.]” Thus, the affidavit sought location history data for an area that encompassed no more than a part of the church and the church grounds, including the parking lot where the assailants waited, a small segment of Veterans Drive between the church and the house at 4923 Veterans Drive, and the house itself, including front and back yards, for a twenty-five minute interval corresponding to the approximate time of the offense.

In a portion of the warrant affidavit explaining “Google Location Services and Relevant Technology[,]” the affiant, Detective Jeffrey Loeb, explained:

Google has developed an operating system for mobile devices, including cellular phones, known as Android, that has a proprietary operating system. Nearly every cellular phone using the Android operating system has an associated Google account, and users are prompted to add a Google account when they first turn on a new Android device. Based on my training and experience, I have learned that Google collects and retains location data from Android-enabled mobile devices when a Google account user has enabled Google location services. Google can also collect location data from non-Android devices if the device is registered to a Google account and the user has location services enabled. The company uses this information for location-based advertising and location-based search results. This location information is derived from GPS data, cell site/cell tower information, and Wi-Fi access points.

In a portion of the affidavit styled “Probable Cause Statement[,]” Loeb next narrated the facts of the offense essentially as described above, concluding with the assertion that:

[i]t is likely that at least one of the four suspects who committed this offense had an Android device on him during the commission of this offense. It is common practice that home invasion robbery suspects keep an open line with someone outside of the residence while committing this type of offense to keep an eye out for responding police officers.

Loeb also averred that he was:

also familiar with Android based cellular devices reporting detailed location information to Google where the electronic data is then stored. This information is captured and recorded even when the user is not doing any specific action on the device. As a result, Affiant is requesting a list of any Google devices in a geographic area around the address of 4923 Veterans Drive, Dallas, Texas 75241 in Dallas County, Texas to help identify the suspects in this capital murder investigation.

The warrant affidavit concluded with a description of the three-step process by which Google releases information in response to geofence warrants, as depicted in the warrant itself and as described above.

C. Execution of the Warrant

The warrant was signed by a district court judge on December 7, 2018.⁶ Pursuant to Step One of the procedure, as outlined in both the warrant and the warrant affidavit, Google identified three devices within the geofence. Once the search was expanded via Step Two, Leob was able to determine that one of those three devices belonged to an individual who was involved in the offense. Step Three revealed that Appellant was that individual. From there, by separate warrants, Loeb was able to obtain Appellant's Google account information plus additional cell phone records to confirm his presence at the crime scene.

D. In the Trial Court

Appellant filed a pretrial motion to suppress evidence obtained pursuant to the geofence warrant. He argued that it constituted an unconstitutional general warrant in that it failed to identify a particular suspect and would thus only serve to invade the privacy of any number of individuals who had nothing to do with the capital murder in this case.⁷ He also argued that the warrant affidavit lacked probable cause to believe any of the assailants were carrying a cell phone with a Google account.⁸

The State responded that, under the circumstances in this case, the Initial Search Parameters were so narrow that “every single device operating in th[e] area,” would have to have been possessed by “either a suspect or a witness.” The prosecutor argued that the geofence warrant was “specifically limited in order to maximize the possibility of returning evidence of a crime and minimize the possibility of intrusion on innocent people.” The trial court ultimately ruled that the warrant affidavit and the warrant itself presented “sufficient particularity to be valid.”

E. In the Court of Appeals

After canvassing the limited authorities (mostly federal cases) that have addressed geofence warrants, the court of appeals concluded:

The geofence warrant cases to date can generally be divided into two categories—those in which the geofence search warrant was found constitutionally infirm because it was not sufficiently limited as to time and place so as to restrict the executing officer's discretion and minimize the danger of searching uninvolved persons, and those in which the warrant satisfied the Fourth Amendment because it established probable cause to search every person found within the geofence area.

Wells, 675 S.W.3d at 826−27. Because “the geofence warrant [in this case] was as narrowly tailored as possible to capture only location data for suspects and potential witnesses[,]” the court of appeals concluded that “the warrant here falls into the second category” as identified in the cases. Id.

Addressing Appellant's argument that the warrant affidavit failed to establish probable cause to believe that any of the suspects were carrying a device with enabled Google location services, the court of appeals invoked the well-known ubiquity of cell phones in modern society. Id. at 826. The court of appeals observed that, “[a]lthough it is possible the suspects were not carrying cell phones with enabled Google location services during the offense, probable cause is about ‘fair probabilities,’ not near certainties.” Id. We agree.

III. Applicable Law

A. Probable Cause and Particularity

The Fourth Amendment provides:

The right of the people to be secure in their persons, houses, papers, and effects, against unreasonable searches and seizures, shall not be violated, and no Warrants shall issue, but upon probable cause, supported by Oath or affirmation, and particularly describing the place to be searched, and the persons or things to be seized.

U.S. Const. amend. IV. As the court of appeals did, Wells, 675 S.W.3d at 827, we will assume (without deciding) that for law enforcement to obtain Google cell phone location history data for a particular area at a particular time constitutes a “search” within the parameters of the Fourth Amendment.⁹

The United States Supreme Court has said that, generally, when law enforcement officers undertake a search for evidence of criminality, before that search may be deemed “reasonable” under the Fourth Amendment, they must first obtain a warrant. Carpenter v. United States, 585 U.S. 296, 316, 138 S.Ct. 2206, 201 L.Ed.2d 507 (2018). Here, a warrant was obtained. The search pursuant to the geofence warrant was therefore reasonable so long as the warrant affidavit supplied probable cause to justify the search, and the warrant itself set out the place to be searched and the things to be seized with sufficient particularity to avoid granting the officers unguided discretion in conducting the search. Dalia v. United States, 441 U.S. 238, 255, 99 S.Ct. 1682, 60 L.Ed.2d 177 (1979); Steagald v. United States, 451 U.S. 204, 220, 101 S.Ct. 1642, 68 L.Ed.2d 38 (1981). See Bonds v. State, 403 S.W.3d 867, 874−75 (Tex. Crim. App. 2013) (listing “limiting the officer's discretion and narrowing the scope of the search” and “minimizing the danger of searching the person or property of an innocent bystander or property owner” as among the objectives of the particularity requirement).

B. The Cases Addressing Geofence Warrants

Geofence warrants are a relatively new phenomenon, having only come into use “since 2016[.]” Note, Geofence Warrants and the Fourth Amendment, 134 Harv. L. Rev. 2508, 2509−10 (May 2021). The few cases so far that have addressed their legitimacy have tended to emanate from lower federal courts and intermediate state appellate courts. And, as the court of appeals observed, those cases “can generally be divided into two categories[.]” Wells, 675 S.W.3d at 826−27.

Which category a given case falls into depends upon the size of the area covered by the requested geofence, the length of time specified, and the circumstances of the offense under investigation. Geofence warrants that are confined, covering a relatively small space over a relatively short time, in a remote or rural area, or at a time of day when only the perpetrators of the offense or witnesses would be likely to be present, have generally been found to pass constitutional muster.¹⁰ But warrants that cover larger or more congested urban areas over a longer span of time generally have not, since they are much more likely to infringe upon a greater number of innocent, uninvolved bystanders.¹¹ Indeed, the issue is often not so much whether there is probable cause to believe the search will uncover evidence of the offense as it is whether the warrant is “overbroad”—that is, whether the search it authorizes outstrips the probable cause that justifies it by casting too wide a net and thereby impacting an unacceptable number of people who cannot possibly have any connection to the offense.¹²

IV. Analysis

In this case, the geofence warrant affidavit supplied ample probable cause to believe both that an offense had occurred and that evidence of the identity of one or more of the perpetrators could be discovered by searching the Google database. Moreover, the warrant itself was framed narrowly enough that almost any device found to have been present within its parameters would have belonged to one of the perpetrators, or potentially to a witness who might identify the perpetrators or testify about the offense, but not merely an innocent bystander.

A. Probable Cause

1. An Offense Occurred ․

Probable cause to support a search warrant is present when, under the totality of circumstances, there is at least a “fair probability” or “substantial chance” (it need not be “more likely than not”) that evidence of an offense will be found at the location that law enforcement seeks to search. Flores v. State, 319 S.W.3d 697, 702 (Tex. Crim. App. 2010) (quoting Illinois v. Gates, 462 U.S. 213, 238, 243 n.13, 103 S.Ct. 2317, 76 L.Ed.2d 527 (1983)).¹³ There can be little doubt—and Appellant does not contest—that the warrant affidavit established that an offense had occurred, namely, the murder of Giddings. The question is whether there is probable cause—at least a “fair probability” or “substantial chance”—that the thing to be searched—the Google location history database—will contain evidence of that offense. For the following reasons, we agree with the court of appeals that the magistrate who signed off on the warrant affidavit had a “substantial basis” to conclude that probable cause existed to believe that Google's location history database would reveal evidence of who murdered Giddings (if it revealed any information at all). Id. (citing Gates, 462 U.S. at 238, 103 S.Ct. 2317).

2. ․ For Which Evidence of the Perpetrator's Identity Could be Found in Google's Location History Database

Detective Loeb's warrant affidavit established that Google's location history database could contain location information for a substantial number of both Android devices as well as non-Android devices that were registered to Google accounts with enabled location services. Appellant argues, however, that probable cause requires a specific showing that one of the assailants was indeed carrying a device with enabled Google location services. Like the court of appeals, we disagree that probable cause necessarily requires so much. Wells, 675 S.W.3d at 826.

Loeb's affidavit claimed that “[i]t is likely that at least one of the four suspects ․ had an Android device on him during the commission of the offense,” since home-invasion-type offenses commonly involve “someone outside of the residence ․ to keep an eye out for responding police officers.” From this the magistrate could reasonably have inferred a “fair probability” or “substantial chance” that the home invaders carried cell phones to keep contact with an outside lookout.

Moreover, a magistrate is entitled to take it as well-established fact that, in this day and age, almost everyone possesses a cell phone on or about his person at practically any time of day or night—they are, indeed, ubiquitous. See Carpenter, 585 U.S. at 311, 138 S.Ct. 2206 (noting that people “compulsively carry cell phones with them at all times”); Riley v. California, 573 U.S. 373, 385, 134 S.Ct. 2473, 189 L.Ed.2d 430 (2014) (noting that cell phones “are now such a pervasive and insistent part of daily life that the proverbial visitor from Mars might conclude they were an important feature of human anatomy”). “The core inquiry here is probability, not certainty, and it is eminently reasonable to assume that criminals, like the rest of society, possess and use cell phones to go about their daily business.” In re Search of Information that is Stored at Premises Controlled by Google LLC, 579 F.Supp.3d 62, 78 (D.D.C. 2021).¹⁴

In this case, the warrant-issuing magistrate had a “substantial basis” to conclude that there was a “fair probability” (or “substantial chance”) that at least one of the four assailants possessed a device that Google could locate within the geofenced area.¹⁵ This constitutes probable cause to believe that Google's location history database would contain evidence relevant to the identity of the person who killed Giddings.

B. Particularity

We also agree that the geofence warrant in this case provided sufficient particularity with respect to both the “place to be searched” and the “things to be seized.” U.S. Const. amend. IV. The “place” designated by the warrant to be searched, which was directed to “Google LLC,” was wherever Google stores its “[r]ecords pertaining to GPS, WiFi or Bluetooth sourced location history data[.]” The “thing to be seized” was the “location history data generated from devices that reported a location within the geographical region bounded by the following latitudinal and longitudinal coordinates, dates and times (“Initial Search Parameters”) and Identifying information for Google Accounts associated with the responsive location history data[.]”

The warrant then identified the specific latitudinal and longitudinal coordinates, narrowly drawn to include no more than a part of the church, the appurtenant church grounds where the assailants waited, the street they rushed across, the front yard of the house where Dickerson was shot, and the house itself, in which Giddings was killed. It also gave a specific date—the date of the murder—as well as the 25-minute window of time during which the offense took place. This degree of specificity appropriately circumscribed police discretion, limiting the information they could obtain from the location history database to that which was relevant to identifying whoever was present at the specific time and place of the offense itself.

Moreover, the “Initial Search Parameters” were sufficiently tailored in terms of time and place as to minimize the potential for infringing on the privacy rights of persons who could not reasonably be regarded as either suspects or witnesses to the offense. The area to be searched was small and restricted to the places where police knew that the unidentified suspects were located: a part of the church grounds, where the suspects hid in waiting for Giddings to arrive home; the street between the church and the home, which the suspects rushed across; and the front yard and interior of the house itself, where the assaults on Dickerson and Giddings took place. These were not high traffic areas—especially not during the brief period of time in the middle of the night when the offense occurred.¹⁶ It was not at all likely that the geofence in this case would have identified many innocent bystanders or passersby who would not have been relevant to the investigation.

Indeed, the “Initial Search Parameters” were so narrow in this case as to allay any concern about Steps Two and Three of the warrant process, as described in both the warrant and the warrant affidavit. Appellant argues that, even if there was probable cause to support Step One of the processes authorized by the geofence warrant in this case, police should have been required to involve the magistrate in deciding which devices identified by Step One merited additional disclosure from the location history database, via Steps Two and Three. Otherwise, Appellant asserts, the warrant allowed police improper unilateral discretion to determine whether there is probable cause to justify that additional disclosure, contrary to Fourth Amendment principles.

But while some geofence warrants may be so broad in scope at Step One as to call for additional involvement by a magistrate at the later stages, we do not believe that to be the case here. The Initial Search Parameters in this case were sufficiently narrow as to provide probable cause, under the circumstances, to believe that whichever devices were revealed to have been present at the narrowly circumscribed place and time captured by the geofence polygon would almost certainly have belonged to legitimate suspects, or potential witnesses, so that any additional disclosure of information via Steps Two and Three would be justified by the same probable cause that supported Step One. See Price v. Superior Court of Riverside County, 93 Cal.App.5th 13, 44−45, 310 Cal.Rptr.3d 520, 546−47 (2023) (concluding that, when the warrant affidavit supplied probable cause “to seize all location data and identifying information for all devices traversing the geofence[,]” additional disclosure in Step Two did not improperly vest police with unguided discretion). On the specific facts of this case, we do not deem the geofence search warrant to have been so lacking in particularity as to require an additional magisterial imprimatur in the later stages of its execution.

V. Conclusion

Assuming that the Fourth Amendment generally requires police to obtain a search warrant for corporate-held location history data, we conclude that the geofence warrant in this case was supported by probable cause and that it satisfied the particularity requirement of the Fourth Amendment. The judgment of the court of appeals is affirmed.¹⁷

CONCURRING OPINION

I agree with the Court's judgment today affirming the judgment of the court of appeals below. I also join Judge Yeary's opinion that explains why the State's geofence warrant was constitutional. Both Judge Yeary and the court of appeals below assumed, without deciding, that law enforcement obtaining cell phone location history data from Google was a “search” under the Fourth Amendment. Notwithstanding my joining Judge Yeary's opinion today that would uphold the constitutionality of the geofence warrant, I write separately to explain that, in my view, we do not need to reach that issue: Law enforcement did not conduct an unreasonable search under the Fourth Amendment because Appellant did not have a reasonable expectation of privacy in the information he voluntarily turned over to a third party.

I. Applicable Law

The Fourth Amendment protects “[t]he right of the people to be secure in their persons, houses, papers, and effects, against unreasonable searches and seizures.” U.S. Const. amend. IV. “A Fourth Amendment privacy interest is infringed when the government physically intrudes on a constitutionally protected area or when the government violates a person's ‘reasonable expectation of privacy.’ ” United States v. Smith, 110 F.4th 817, 830 (5th Cir. 2024) (quoting United States v. Jones, 565 U.S. 400, 406, 132 S.Ct. 945, 181 L.Ed.2d 911 (2012)). To determine whether a “reasonable expectation of privacy exists,” Justice Harlan's two-step approach articulated in his concurring opinion in Katz v. United States, 389 U.S. 347, 88 S.Ct. 507, 19 L.Ed.2d 576 (1967), controls. See Jones, 565 U.S. at 406, 132 S.Ct. 945. A defendant must show (1) that he had a subjective expectation of privacy; and (2) that his subjective expectation of privacy is one that society recognizes, or is prepared to recognize, as reasonable. Smith v. Maryland, 442 U.S. 735, 740, 99 S.Ct. 2577, 61 L.Ed.2d 220 (1979).

The Supreme Court of the United States has applied the reasonable expectation of privacy test to electronic information in several cases. In United States v. Knotts, 460 U.S. 276, 103 S.Ct. 1081, 75 L.Ed.2d 55 (1983), for example, the Court held that the use of “beeper” information to track a vehicle's movements was not a Fourth Amendment search. 460 U.S. at 281, 103 S.Ct. 1081. The Court emphasized that the movements of Knott's vehicle and its final destination had been voluntarily conveyed to anyone who wanted to look, and therefore he could not assert a privacy interest in the information obtained. Id. Three decades later, the Court decided United States v. Jones, 565 U.S. 400, 132 S.Ct. 945, 181 L.Ed.2d 911 (2012). There, the Court addressed whether the remote monitoring of a vehicle's movements via an attached GPS tracking device for twenty-eight days violated a person's legitimate expectation of privacy. 565 U.S. at 402–04, 132 S.Ct. 945. Applying a physical-trespass theory (instead of Katz’s expectation-of-privacy analysis), a majority of the Court said yes. Id. at 410–11, 132 S.Ct. 945.

Other cases are equally instructive. In United States v. Miller, 425 U.S. 435, 96 S.Ct. 1619, 48 L.Ed.2d 71 (1976), the Court held that the government did not conduct a search when it obtained an individual's bank records from his bank, since he voluntarily exposed those records to the bank in the ordinary course of business. 425 U.S. at 443, 96 S.Ct. 1619. Finally, in Smith, the Court held that the government did not conduct a search when it used a pen register to record outgoing phone numbers dialed from a person's telephone, because he voluntarily conveyed those numbers to his phone company when placing calls. 442 U.S. at 742, 99 S.Ct. 2577.

Then came Carpenter v. United States, 585 U.S. 296, 138 S.Ct. 2206, 201 L.Ed.2d 507 (2018). In Carpenter, the Court addressed whether a person has a legitimate expectation of privacy in historical cell-site location information (CSLI) records. Id. at 303, 138 S.Ct. 2206. The Court concluded that the defendant had a reasonable expectation of privacy in his historical CSLI, and the government violated the Fourth Amendment when it searched the location records without a warrant supported by probable cause. Id. at 316–17, 138 S.Ct. 2206. Critically, the Court emphasized that the defendant had not voluntarily turned over his CSLI to a cell phone provider as understood in Miller and Smith. Id. at 309, 138 S.Ct. 2206.

This Court has addressed CSLI in several cases. Sims v. State, 569 S.W.3d 634 (Tex. Crim. App. 2019), was the first time, and this Court held that an individual “did not have a legitimate expectation of privacy in his physical movements or his location as reflected in the less than three hours of real-time CSLI records accessed by police by pinging his phone less than five times.” 569 S.W.3d at 646. A year later, in Holder v. State, 595 S.W.3d 691 (Tex. Crim. App. 2020), this Court held that accessing “23 days” of CSLI violated Article I, Section 9 of the Texas Constitution. 595 S.W.3d at 704. Prior to Carpenter, this Court also addressed searches of CSLI records in Ford v. State, 477 S.W.3d 321 (Tex. Crim. App. 2015), and Love v. State, 543 S.W.3d 835 (Tex. Crim. App. 2016).¹

Turning to Google's location history data, in United States v. Chatrie, 107 F.4th 319 (4th Cir. 2024), the Fourth Circuit held that the government “did not conduct a search when it obtained [the location history] data” of an individual. 107 F.4th at 332. The Fourth Circuit first distinguished location history data from CSLI, noting that location history data is not an “all-encompassing record of whereabouts,” rather, the information is “far less revealing” and limited to a “single, brief trip.” Id. at 330. Next, the Fourth Circuit noted that the appellant “voluntarily exposed his location information to Google” and did so knowingly, having been warned of the consequences when he opted into Google's services. Id. at 331. Thus, the Fourth Circuit concluded that the third-party doctrine governed the case, so the appellant could not claim a reasonable expectation of privacy in the information voluntarily exposed to Google. Id.

The Fifth Circuit soon after disagreed. In United States v. Smith, 110 F.4th 817 (5th Cir. 2024), the Fifth Circuit held that geofence warrants “are general warrants categorically prohibited by the Fourth Amendment.” 110 F.4th at 838. The Fifth Circuit rejected Chatrie and held that “geofence location data is invasive for Fourth Amendment purposes.” Id. at 834. For the Fifth Circuit, the question was whether location history data “ha[d] the capability of revealing intimate, private details about a person's life,” not whether the search itself did. Id. at 834 n.8. The Fifth Circuit then held that the third-party doctrine did not apply because, while an individual does “opt in” to the services, the decision to opt in is “hardly informed and, in many instances, may not even be voluntary.” Id. Concluding that a search had occurred, the Fifth Circuit then held that geofence warrants were unconstitutional general warrants because the actual database search (at Step 1 of the process), involved a search of the entire Google repository—as the Fifth Circuit described it, “general, exploratory rummaging.” Id. at 837. Nevertheless, applying the good-faith exception to the exclusionary rule, the Fifth Circuit affirmed the district court's denial of the motion to suppress. Id. at 840.²

II. Analysis

a. Google's Location History Data is not CSLI.

Carpenter described CSLI as “an all-encompassing record of the [device] holder's whereabouts,” that “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.” Carpenter, 585 U.S. at 311, 138 S.Ct. 2206 (citing Jones, 565 U.S. at 415, 132 S.Ct. 945) (internal quotations omitted). A cell phone “tracks nearly exactly the movements of its owner ․ faithfully follow[ing] ․ beyond public thoroughfares and into private residences, doctor's offices, political headquarters, and other potentially revealing locales.” Id. Since a person has a reasonable expectation of privacy in the whole of their physical movements, it follows that location information obtained from the tracker in one's pocket via wireless carriers was the product of a search.

But location history data is noticeably different. A geofence warrant reveals no more than “an individual trip viewed in isolation.” Chatrie, 107 F.4th at 330 (quoting Beautiful Struggle v. Balt. Police Dep't, 2 F.4th 330, 342 (4th Cir. 2021) (en banc)). The information obtained is more akin to the short-term public movements in Knotts, or the short-term CSLI records this Court analyzed in Sims, rather than the records in Carpenter and Jones. I cannot say that Appellant had a legitimate expectation of privacy in his location records for the limited timeframe the State sought in this case.

b. The information was voluntarily exposed.

As the Supreme Court correctly recognized in Carpenter, “[c]ell phone location information is not truly ‘shared’ as one normally understands the term.” 585 U.S. at 315, 138 S.Ct. 2206. Importantly, “a cell phone logs a cell-site record by dint of its operation, without any affirmative act on the part of the user beyond powering up.” Id. “[A]part from disconnecting the phone from the network, there is no way to avoid leaving behind a trail of location data.” Id. Thus, “in no meaningful sense does the user voluntarily assume the risk of turning over a comprehensive dossier of his physical movements.” Id. (citing Smith, 442 U.S. at 745, 99 S.Ct. 2577).

Not so, here. As the record in this case makes clear, when a user does not affirmatively turn on the location history settings, a mobile device's location will not be automatically saved. Appellant undertook affirmative steps on his Android device to enable location sharing: He logged in to his Google account on the device and opted in to the Location History services in the account's settings. He also had to click through several warning screens that admonished him of the consequences of opting in. Those consequences included, as relevant here, Google collecting and tracking Appellant's location history and sharing that data with additional third-parties, inter alia, advertisers.³ After providing consent to Google, Appellant still retained ultimate control over both his settings (he could limit or disable Location History at any time), and the data collected by Google (he could edit or delete any location history data through his Timeline). At any time, Appellant had the ability to cease sharing his location history information with Google and withdraw his consent for Google to share that information with other third parties. Thus, Appellant voluntarily exposed his location to Google. The Fourth Circuit in Chatrie put it best: “If Google compiles a record of [a user's] whereabouts, it is only because he has authorized Google to do so.” 107 F.4th at 331. I cannot say that Appellant had a reasonable expectation of privacy in information he voluntarily turned over to a third party.

III. Conclusion

I agree with the Court's judgment to affirm the judgment of the court of appeals. I only write separately to express my view that Appellant did not have a reasonable expectation of privacy in information that he voluntarily turned over to Google. Law enforcement did not need a warrant to obtain that information. With these thoughts, I join the Court's judgment.

The most important question before us is whether the geofence warrant amounts to a constitutionally protected search. Instead of answering that question, Judge Yeary's opinion assumes it away and in doing so crafts an opinion that ensures that we will never have to answer the question. It turns this case from a geofence warrant case into a probable cause case that will essentially lower the standard for probable cause for all warrants just to uphold a search pursuant to a novel type of warrant.

I disagree that the geofence warrant in this case was adequately supported by probable cause. I would hold instead that Appellant did not have a legitimate expectation of privacy in the limited information sought through the geofence warrant's first and second steps. These two steps sought temporally and spatially limited location history data consistent with existing case law regarding cell phone location information.

But I would hold that Appellant did have a reasonable expectation of privacy in the information sought by the warrant's third step, which included six months of prior IP history.¹ Therefore, I concur with upholding at least the first two steps authorized by the geofence warrant but not on the basis of probable cause. But because the court of appeals did not address the required threshold question of whether Appellant had a reasonable expectation privacy in the information sought by the geofence warrant I dissent to refusal to do so on that point.

Was There a Constitutionally Protected Search?

The Fourth Amendment's “basic purpose ․ is to safeguard the privacy and security of individuals against arbitrary invasions by government officials.”² “[W]hen an individual ‘seeks to preserve something as private,’ and his expectation of privacy is ‘one that society is prepared to recognize as reasonable,’ official intrusion into that sphere generally qualifies as a search and requires a warrant supported by probable cause.”³ We have recognized that “[t]he threshold issue in every Fourth Amendment analysis is whether a particular government action constitutes a ‘search’ or a ‘seizure.’ ”⁴ The State argued at trial, on appeal, and argues before this Court that Appellant has failed to establish a search occurred because he does not have a reasonable expectation of privacy in his location history data. Before reaching the question of probable cause, the court of appeals should have answered this threshold question.⁵

A common thread in the jurisprudence regarding the expectation of privacy is that voluntarily sharing things with others generally defeats an objective expectation of privacy. For example, in United States v. Knotts, the Supreme Court considered law enforcement's use of a planted beeper's signal to track a vehicle through traffic and recognized that movement in an otherwise public area is not protected by an expectation of privacy.⁶ Similarly, when it comes to bank records, the turning over of financial information to a third party has been seen as a voluntary relinquishment of an expectation of privacy in the bank records themselves.⁷ But the Court has also recognized that temporal limits should nevertheless be placed upon otherwise voluntarily disclosed information. For example, tracking information for an extended period of time even in a public place might infringe upon an expectation of privacy.⁸

From these cases it appears that seizing arguably voluntarily disclosed location information does not infringe upon an expectation of privacy if the information sought is both spatially and temporally limited. In Carpenter, the Court held there was a reasonable expectation of privacy in at least seven days of historical cell-site location information (“CSLI”) associated with Carpenter's phone and, as a result, the Fourth Amendment was violated when the phone was searched without a warrant supported by probable cause.⁹ The Court reasoned that cell-site location records hold “the privacies of life” by revealing “not only particular movements, but through them [a person's] ‘familiar, political, professional, religious, and sexual associations.’ ”¹⁰ The Court held that “[g]iven the unique nature of cell phone location records, the fact that the information is held by a third party does not by itself overcome the user's claim to Fourth Amendment protection.”¹¹ Post-Carpenter, we recognized that, “[w]hether a particular government action constitutes a ‘search’ or ‘seizure’ does not turn on the content of the CSLI records; it turns on whether the government searched or seized ‘enough’ information that it violated a legitimate expectation of privacy.”¹² In Sims v. State, this Court was considering real-time location information as opposed to historical CSLI but we found the reasoning in Carpenter applicable to both kinds of records.¹³ And we held that Sims did not have a legitimate expectation of privacy in less than three hours of real-time CSLI records accessed by law enforcement pinging his phone less than five times.¹⁴

Geofence warrants, like the one at issue, have been said to “work in reverse” to the traditional search warrant.¹⁵ Geofence warrants generally specify “a location and period of time, and, after judicial approval, companies conduct sweeping searches of their location databases and provide a list of cell phones and affiliated users found at or near a specific area during a given timeframe, both defined by law enforcement.”¹⁶ Here, “[i]n the first step, Google would be asked to create an anonymized list of all devices located within the ‘target location’ during the time period of 2:45 a.m. to 3:10 a.m. on June 24, 2019.”¹⁷ The target location was “limited to the house where the offense occurred and a portion of the church property across the street.”¹⁸ The target location primarily sought location history data from public spaces.¹⁹ With that information, law enforcement would “analyze this location data to identify users who may have witnesses or participated” in the offense. Google “would then provide additional anonymized location history outside of the target location for a period not to exceed sixty minutes before and after the last timestamp associated with the device within the target location.”²⁰ The first step was limited to a 25-minute duration, during the time of the offense, at the location of the offense. The second step expanded the location but was still limited temporally. The first two steps involved less than three hours of location information. Given these temporal and spatial limitations, Appellant did not have a reasonable expectation of privacy in the location history data returned. The State did not seek “enough” information through these steps that it can be said to have violated a legitimate expectation of privacy.²¹ This limited intrusion did not implicate the type of “privacies of life” that were at the center of the Court's concern in Carpenter.²²

There is another relevant distinction between the CSLI records at issue in Carpenter and the location history information with which we are presently concerned. The Court in Carpenter found the CSLI records had not been voluntarily “shared” in the way that term is normally understood.²³ The Court noted the pervasive nature of cell phones in modern society, that cell-site records are created without any affirmative action on the part of the user given that virtually activity on the phone generates CSLI, and concluded that, “in no meaningful sense does the user voluntarily ‘assume[ ] the risk’ of turning over a comprehensive dossier of his physical movements.”²⁴ Here, the Google location history data was recorded by Google only if users “opted-in.” There was an affirmative action by the user rather than a recording of location information caused by virtually any activity on the phone.²⁵ Arguably this step suggests a greater assumption of the risk than merely using a cell phone, but I question whether turning on location services to be able to use a particular app or even ask your phone for directions to a particular location is the same thing as agreeing to be surveilled for an extended period of time. Still, in this case, the information produced in the first two steps of the geofence warrant was not so open ended that it sought a dossier of physical movements. Rather, it captured a device's presence in a discrete location for a short amount of time.

The voluntariness of the disclosure and the limited scope of the data sought support the application of the third-party doctrine, which holds that “a person has no legitimate expectation of privacy in information he voluntarily turns over to third parties.”²⁶ The Court in Carpenter declined to apply the third-party doctrine to the CSLI sought concluding that the rationales underlying the doctrine did not support its application because (1) the information was not voluntarily shared and (2) the nature of the information sought.²⁷ Here, both rationales support the application of the third-party doctrine as to the limited location history data and subscriber information sought. For those records sought in the first two steps, this is not the “rare case” where there is a legitimate privacy interest in the records held by a third party.²⁸

But the third step of the geofence warrant sought six months of prior IP history in addition to identifying subscriber information. Judge Yeary acknowledges that the information provided in step three “may suggest access” to more than identifying information. We have previously recognized that a person has a legitimate expectation of privacy in the contents of his cell phone.²⁹ Unlike brief and limited location history data or subscriber information, months of IP history data has the ability to reveal the “privacies of life.”³⁰ The rationales underlying the third-party doctrine do not hold up when considering months of IP history data. First, the nature of that data is not limited in time or scope and are potentially much more revealing than limited location history data. Furthermore, there is no “opt-in” for the record kept of websites visited or services accessed on a cellphone. Cellphones log IP history whenever the user accesses the internet. This data, like 127 days of CSLI, “provides an intimate window into a person's life,” which can provide location as well as “familial, political, profession, religious, and sexual associations.”³¹ When the information sought is not temporally or spatially limited, implicates privacy concerns, and is not voluntarily shared, the third-party doctrine does not overcome an expectation of privacy. The geofence warrant ultimately sought months of historical data in which Appellant had a reasonable expectation of privacy. Thus, I would hold, at a minimum that the third step of the warrant constituted a constitutionally protected search. To clarify, I am not suggesting that the existence of a legitimate expectation of privacy in the IP address information invalidates the entire warrant. Rather, I would hold that the warrant must be based upon probable cause in light of the third step authorized by the warrant. And, as I discuss below, I do not agree that there is probable cause supporting any of the steps authorized by the warrant, especially the third.

Probable Cause

Today, however, the Court does not decide whether the information sought by the geofence warrant constitutes a search. Rather, it decides the case on the basis of probable cause. In doing so, it virtually ensures that the threshold question will never be answered. Judge Yeary's opinion effectively lowers the burden for law enforcement to obtain a warrant based upon probable cause given the meager amount of information available to police when this geofence warrant was sought. Even a CSLI case starts with a suspect, but here, law enforcement only had a location.

Given that, I disagree with the court of appeals’ conclusion that the warrant at issue was supported by probable cause. To establish probable cause, an affidavit in support of a search warrant must provide a substantial basis for concluding that there is a fair probability that contraband or evidence of a crime will be found in a particular location.³² Here, the affidavit asserted it was likely that one of the suspects carried an Android phone and that home invasion suspects commonly communicate with someone outside of the residence but provided no other basis for concluding evidence of a crime could be found by searching the Google database. In State v. Baldwin, this Court held that in order to conduct a search of the contents of a suspect's cell phone, law enforcement must be able to demonstrate probable cause in the form of a “nexus” between the cell phone and the commission of the offense.³³ Conclusory allegations or conclusions are generally insufficient to establish probable cause.³⁴ Applying that understanding to boilerplate language about the use of cellphones among criminals, the Court held that “specific facts connecting the items to be search to the alleged offense are required for the magistrate to reasonably determine probable cause.”³⁵ Here, there were no specific facts connecting the Google location history database to the alleged offense beyond a conclusory statement about the likelihood a suspect carried an Android phone. There was no basis for that conclusion.

In Baldwin, witnesses identified Baldwin's vehicle leaving the victim's home the day of the murder.³⁶ Baldwin was later stopped in the sedan matching the witnesses’ description and a cellphone was located inside of the vehicle.³⁷ Investigators obtained a warrant to search the phone. The warrant's affidavit noted that available geolocation information may show the location of the suspect at or near the time of an offense and that the investigator knows from “training and experiences that someone who commits the offense of aggravated assault or murder often makes phone calls and/or text messages immediately prior and after the crime.”³⁸ The Court rightly held this was not sufficient to establish probable cause. While we recognized that the witnesses’ descriptions and the vehicle's license plate supported a nexus between the vehicle and the offense, “they ha[d] no bearing on whether [Baldwin's] phone [wa]s connected with the offense” because “[t]he affidavit contains nothing about the phone being used before or during the offense.”³⁹ We reiterated that “[s]uspicion and conjecture do not constitute probable cause.”⁴⁰ Without more, we held, boilerplate language about cellphone use among criminals is insufficient to establish probable cause.⁴¹

Here, the “nexus” provided in the affidavit to connect the cellphone, and the criminal offense was that “[i]t is likely that at least one of the four suspects who committed this offense had an Android device on him during the commission of this offense. It is common practice that home invasion robbery suspects keep an open line with someone outside of the residence while committing this type of offense to keep an eye out for responding police officers.” This amounts to nothing more than boilerplate language about the use of cellphones among criminals.⁴² Without more, this is the same type of bare conclusion or speculation about the use of cellphones that does not support probable cause. Nothing in the affidavit suggests, for example, that the phones at issue were anything more than at the target location. There is nothing suggesting they were used during the commission of the offense or captured information regarding the crime.⁴³

Judge Yeary concludes that from the affidavit's assertions that “[i]t is likely that at least one of the four suspects ․ had an android” and that it is common in-home invasions for suspects to “keep an open line” with a lookout outside of the residence, a magistrate could have reasonably inferred a “fair probability” that the home invaders carried cellphones. But “a magistrate's action cannot be a mere ratification of the bare conclusions of others.”⁴⁴ In other words, the affidavit for the search warrant itself cannot simply assert that there is a fair probability or likelihood and that alone be considered sufficient to support a finding of probable cause. Such conclusions must be supported by facts. Even if the magistrate could accept the conclusion offered, the warrant does not include any facts to establish a fair probability that the devices identified were related to the offense.⁴⁵ There is nothing more than boilerplate language about the use of cellphones in home invasions.

Judge Yeary's opinion allows for the idea that anyone identified via a device located within the search warrant parameters would necessarily be a suspect or a witness. But by definition a witness is not involved in the crime. Like a general warrant, searching for the identity of a witness presupposes there is no evidence of guilt, and it uses the search itself to generate evidence to test that witness involvement.⁴⁶ This is the precise danger created by the way in which a geofence warrant works backwards to gather evidence. There is nothing but speculation to suggest that a potential witness had any information on their phone regarding the crime. Introducing the idea of probable cause to believe that a “witness” has evidence of a crime significantly lowers the specificity required to obtain a warrant.⁴⁷ It fails to distinguish between innocent bystanders and suspects. Indeed, one of the people at issue was excluded after this evidence was obtained.⁴⁸

Judge Yeary argues probable cause is supported by the fact that, today, almost everyone possesses a cell phone on his person. Indeed, the Supreme Court in Carpenter recognized that people “compulsively carry cell phones with them all the time.”⁴⁹ And the Court previously characterized cell phones as pervasive.⁵⁰ But the Supreme Court's acknowledgment of the ubiquitous nature of cell phones was made to acknowledging greater privacy concerns for the data captured by cell phones not as support for greater intrusions.⁵¹ In any event, the fact that most people carry cellphones does not provide the required nexus to the offense. That most people carry cellphones without more does not establish a fair probability that evidence of a crime will be discovered by searching phones located near the scene of a crime.

Conclusion

The warrant in this case did not establish probable cause. But because the court of appeals erred in the first instance by failing to consider whether Appellant had a reasonable expectation of privacy in the things to be searched, I would remand for the court of appeals to consider that question. In particular, I believe the court of appeals should consider whether the third step in the geofence warrant amounted to an intrusion into a legitimate expectation of privacy. I would require the court of appeals to reconsider whether probable cause could justify the search of six months of IP address information under that warrant, and, if necessary, whether the second warrant in this case was obtained as a fruit of the poisonous tree.

FOOTNOTES

1.   Succinctly put, geofence warrants have been described in this way:While traditional court orders permit searches related to known suspects, geofence warrants are issued specifically because a suspect cannot be identified. Law enforcement simply specifies a location and period of time, and, after judicial approval, companies conduct sweeping searches of their location databases and provide a list of cell phones and affiliated users found at or near a specific area during a given timeframe, both defined by law enforcement.Note, Geofence Warrants and the Fourth Amendment, 134 Harv. L. Rev. 2508, 2509 (May 2021).

2.   The first ground of review we granted is: “[w]hether the Court of Appeals correctly determined the legality of geofence warrants, an issue of first impression in Texas and an important question of state and federal law that has not been, but should be, settled by the Court of Criminal Appeals.”

3.   The warrant purported to issue pursuant to former Article 18.21, Section 5A, of the Texas Code of Criminal Procedure. That statute was repealed in 2017, but the repeal was not effective until January 1, 2019. See Acts 2017, 85th Leg., ch. 1058, §§ 5.01(2), 6.03, pp. 4192−93, eff. Jan. 1, 2019. The warrant issued on December 7, 2018. There is no issue before us whether the warrant was properly issued pursuant to statutory authority.

4.   Courts that have addressed geofence warrants refer to these “Step Three” records simply as identifying information. See, e.g., United States v. Smith, 110 F.4th 817, 825 (5th Cir. 2024) (emphasis added) (“[A]t Step 3, law enforcement compels Google to provide account-identifying information[.]”); United States v. Chatrie, 107 F.4th 319, 324 (4th Cir. 2024), reh'g granted en banc, 2024 WL 4648102 (4th Cir. Nov. 1, 2024) (emphasis added) (“[A]t Step Three, law enforcement determines which individuals are relevant to the investigation and then compels Google to provide their account-identifying information[.]”); Price v. Superior Court of Riverside County, 93 Cal.App.5th 13, 22, 310 Cal.Rptr.3d 520, 529 (2023) (emphasis added) (“Geofence warrants allow law enforcement agencies to identify suspects and witnesses to crimes by obtaining location data and identifying information[.]”). Judge Newell worries that the warrant may have gone too far to authorize a search of Appellant's IP history. Concurring and Dissenting Opinion at ––––, –––– – ––––, ––––. There is no suggestion in the record that police actually obtained Appellant's IP history pursuant to this warrant. See Jones v. State, ––– Ga. ––––, ––– S.E.2d ––––, –––– n.5, No. S24A1085, 2025 WL 676862, at *8 n.5 (2025) (“[E]ven if the warrant's broader description of items to be seized might raise concerns about particularity, it would not invalidate the warrant here, because the police neither obtained nor used any evidence beyond what was needed to identify Jones.”). In any event, to the extent that the information provided at Step Three in this case may suggest access to anything other than identifying information, that argument has not been raised in either the court of appeals or this Court, and we will not address it.

5.   Both the affidavit and the warrant itself, besides providing a verbal description of the area included within the geofence, also incorporated a graphic representation of the polygon, which the court of appeals reproduced in its opinion. Wells, 675 S.W.3d at 822.

6.   As “Grounds for Issuance[,]” the warrant affidavit cited Articles 18.02(10) (evidentiary search warrants) and 18.02(13) (electronic customer data held in electronic storage) of the Texas Code of Criminal Procedure. Tex. Code Crim. Proc. arts. 18.02(10), 18.02(13). Former Article 18.21, Section 5A(b), authorized only district court judges to issue the latter type of warrant.

7.   At a hearing on the motion to suppress, counsel for Appellant argued:The geofence warrant in this case did not identify [Appellant] in any way. In fact, it did not identify anyone. Instead, the warrant operated in reverse. It required Google to identify a large cache of deeply private data and then allowed police the discretion to sift through it and obtain private information from devices of interest. * * * The process effectively filtered out the innocent through increasing levels of searches. But such a process illustrates that the searchers themselves knew that they were searching the innocent merely because they walked or drove through an area in which a crime was committed.

8.   At the hearing, counsel for Appellant maintained:[I]t is not enough to submit an affidavit stating that probable cause exists for a geofence warrant because, given broad cell phone useage [sic], it is likely the criminal suspect had a cell phone. If this were the standard, a geofence warrant could issue at almost any criminal investigation where a suspect is unidentified.

9.   The Fourth Circuit Court of Appeals and the Fifth Circuit Court of Appeals have recently disagreed on the question of whether the Government's acquisition of location history data by way of a geofence warrant constitutes a “search” for Fourth Amendment purposes. Compare Chatrie, 107 F.4th at 332 (applying the third party doctrine to hold that it does not constitute a search), with Smith, 110 F.4th at 836 (holding that the third party doctrine does not apply to geofence warrants and concluding that “law enforcement in this case did conduct a search when it sought Location History data from Google”); Pierre Grosdidier, Courts Are Split: A Look at the Constitutionality of Geofence Warrants, 87 Tex. B.J. 776 (Nov. 2024). The court of appeals in this case found it unnecessary to address this question. Wells, 675 S.W.3d at 827. In Price v. Superior Court of Riverside County, a case upon which the court of appeals relied heavily, the intermediate California appellate court likewise “assumed for purposes of discussion ․ that the search for location data and identifying information, as authorized by the geofence warrant, constituted a ‘search’ within the meaning of the Fourth Amendment.” 93 Cal.App.5th at 37 n.9, 310 Cal.Rptr.3d at 541 n.9. We will too.Judge Newell contends that the court of appeals in this case “erred” not to reach the threshold question of whether there was even a search for Fourth Amendment purposes. Concurring and Dissenting Opinion at –––– & n.5, ––––. Even if this Court were to find that the court of appeals erred in its probable cause analysis, however, that would not mean it erred in failing to reach the threshold “reasonable expectation of privacy” question. See Byrd v. United States, 584 U.S. 395, 411, 138 S.Ct. 1518, 200 L.Ed.2d 805 (2018) (“Because Fourth Amendment standing is subsumed under substantive Fourth Amendment doctrine, it is not a jurisdictional question and hence need not be addressed before addressing other aspects of the merits of a Fourth Amendment claim.”); Jones, ––– S.E.2d at –––– n.1, 2025 WL 676862, at *3 n.1. It would just mean that it would be necessary to remand the case to that court, which is what Judge Newell advocates anyway. Concurring and Dissenting Opinion at ––––.

10.   See, e.g., In re Search Warrant Application for Geofence Location Data Stored at Google Concerning an Arson Investigation (“Arson”), 497 F.Supp.3d 345, 358 (N.D. Ill. 2020) (“Streets in the wee hours of the morning in the City of Chicago are generally sparsely populated by pedestrians, and roads have few cars traversing through them. Furthermore, the affiant has provided additional information obtained through the investigation to support the conclusion that location data from uninvolved individuals will be minimized.”); In re Search of Information that is Stored at Premises Controlled by Google LLC, 579 F.Supp.3d 62, 85−86 (D.D.C. 2021) (“[T]he geofence drawn here is located in an industrial area, not a ‘congested urban area,’ and no residences can be seen [within it, and] the target area is small and lightly trafficked enough to render the search reasonable.”); In re Search of Information that is Stored at Premises Controlled by Google, No. 2:22-mj-01325, 2023 WL 2236493, at *12 (S.D. Tex. Feb. 14, 2023) (“The timing of the request indicates narrow tailoring to avoid collection of data at times when uninvolved devices, and the people possessing them, would likely be within the polygon. * * * The polygon itself is also narrowly tailored to ensure that Location History information, with a fair probability, will capture evidence of the crimes only.”); Price, 93 Cal.App.5th at 43−44, 310 Cal.Rptr.3d at 546 (“The target location was limited to the front yard ․ where the shooting occurred, and the street in front of the house, for the length of two houses in each direction, where the two suspects were seen fleeing after the shooting. * * * Additionally, because the warrant sought first-stage location data after 10:00 p.m. in a suburban, residential neighborhood, it was likely that any individuals traversing the geofence were either suspects or witnesses to the shooting.”); Tomanek v. State, 261 Md.App. 694, 715, 314 A.3d 750, 762 (2024) (“[B]y limiting the search area to within a 100-meter radius of the main residence [of a rural homestead], the police virtually ensured that any cell phone activity that met the [geofence] search's parameters would have had to come from within the property's boundaries. Given that the property was privately owned and included ‘no trespassing’ signs, and given that the property owner had claimed no family member had been to the property between April 4 and April 11, 2020, the chance that the search would result in any unauthorized or unnecessary invasion of privacy was almost non-existent.”); State v. Contreras-Sanchez, 5 N.W.3d 151, 168 (Minn. Ct. App. 2024), rev. granted (May 29, 2024) (“The geofence did not include any buildings at all. The closest residence was over 1,200 feet away. And due to the remoteness of the area, the warrant's inclusion of a scarcely used road did not risk pulling in vast swaths of location-history data from drivers who just happened to be passing through this rural area.”); Jones, ––– S.E.2d at ––––, 2025 WL 676862, at *7 (“[T]he time range matched the approximate time period when the suspect was seen at and around the [murder] victim's home, and the geographic range was reasonably targeted to capture the suspect's movements, especially given that Google location history is not precise.”); see also, 2 Wayne R. LaFave, Search and Seizure: A Treatise on the Fourth Amendment § 4.6(d), at 35−37 (6th ed. 2020) (Supp. 2024) (discussing Arson and Price approvingly); Mary D. Fan, Big Data Searches and the Future of Criminal Procedure, 102 Tex. L. Rev. 877, 886 (April 2024) (“[D]igital probable cause and particularity can be established by a tightly framed ․ geofence warrant likely only to net persons for whom there is probable cause to believe perpetrated an unsolved crime.”).

11.   See, e.g., In re Search of Information Stored at Premises Controlled by Google, 481 F.Supp.3d 730, 756 (N.D. Ill. 2020) (finding the geofence warrant overbroad, while also observing: “It is also possible to imagine other applications of geofence technology that might comport with Fourth Amendment standards. Say, for example, that the government develops information supporting probable cause to believe that its geofences will not capture the information of uninvolved persons, such as a scenario in which the government can establish independently that only the suspected offender(s) would be found in the geofence, or where probable cause to commit an offense could be found as to all present there.”); In re Search of Information that is Stored at Premises Controlled by Google, LLC, 542 F.Supp.3d 1153, 1158 (D. Kan. 2021) (“[T]he geofence boundary appears to potentially include the data for cell phone users having nothing to do with the alleged criminal activity. The boundary encompasses two public streets, so anyone driving their automobile by the target location during the relevant time period could be identified in the data. Google Maps also indicates that the subject building contains another business, which the application does not address. * * * And the nexus between the alleged criminal activity and [the] one-hour duration [of the requested temporal scope of the search] is weak.”); People v. Meza, 90 Cal.App.5th 520, 312 Cal.Rptr.3d 1, 18−19 (2023) (“The failure to sufficiently narrow the search parameters potentially allowed a location-specific identification of thousands of individuals ․ for whom no probable cause existed. * * * The warrant here, authorizing the search of more than 20 acres total over a cumulative period of more than five hours in residential and commercial areas did not meet the fundamental threshold requirement [of particularity].”); see also, In re Search of: Information Stored at Premises Controlled by Google, as Further Described in Attachment A, No. 20 M 297, 2020 WL 5491763, at *7 (N.D. Ill. July 8, 2020) (finding the geofence warrant overbroad, but observing that, “if the government had constrained the geographic size of the geofence and limited the cellular telephone numbers for which agents could seek additional information to those numbers that appear in all three defined geofences, the government would have solved the issues of overbreadth and lack of particularity”).

12.   E.g., Meza, 312 Cal.Rptr.3d at 16, 20 (observing that “the warrant in this case sufficiently described the place to be searched (Google's database of users’ location history) and items to be retrieved from that search (designated records for users to be found within the boundaries of certain coordinates at certain times)[,]” while nevertheless concluding that the warrant lacked particularity for failing to be structured so as to minimize the potential for capturing location data for uninvolved individuals and maximize the potential for capturing location data for suspects and witnesses); United States v. Chatrie, 590 F.Supp.3d 901, 929−30 (E.D. Va. 2022) (“To be sure, a fair probability may have existed that the Geofence Warrant would generate the suspect's location information. However, the warrant, on its face, also swept in unrestricted location data for private citizens who had no reason to incur Government scrutiny.”).

13.   Assuming that the warrant in this case was controlled by Article 18.01(c) of the Texas Code of Criminal Procedure, because it is a so-called “evidentiary” search warrant under Article 18.02(10), it could not have issued absent probable cause as to three things: “(1) that a specific offense has been committed, (2) that the specifically described property or items that are to be searched for or seized constitute evidence of that offense or evidence that a particular person committed that offense, and (3) that the property or items constituting evidence to be searched for or seized are located at or on the particular person, place, or thing to be searched.” Tex. Code Crim. Proc. arts. 18.01(c), 18.02(10). For reasons we describe in the text, all three statutory criteria are satisfied here.

14.   In State v. Baldwin, 664 S.W.3d 122, 123 (Tex. Crim. App. 2022), this Court concluded that, in order to conduct a search of the content of a suspect's cell phone, law enforcement must be able to demonstrate probable cause in the form of a “nexus” between the cell phone and the commission of the offense itself. The Court later concluded that probable cause can be established that the contents of a cell phone might contain evidence of a crime without necessarily showing that the cell phone was directly used in the commission of the offense. Stocker v. State, 693 S.W.3d 385, 388 (Tex. Crim. App. 2024). In any event, in the instant case, mere possession of the cell phone at the time and location of the offense may well be enough—never mind what is contained in the cell phone—to constitute evidence of the identity of the perpetrator of, or witnesses to, the offense.

15.   One legal commentator has observed that, at least as of 2022:[S]tatistics demonstrate a fair probability that Google will have location data on the target of a geofence query. Eighty-five percent of Americans own a smartphone, and Google enjoys a 40% domestic market share (compared to 60% for Apple). This means that 34% of Americans own an Android smartphone while 51% own in iPhone. If Google has location data on almost all Android users (30%) and half of iPhone users (25%), then Google has location information on 55% of Americans, meeting the preponderance standard, and thus by definition satisfying the lower probable cause test.Reed Sawyers, For Geofences: An Originalist Approach to the Fourth Amendment, 29 Geo. Mason L. Rev. 787, 807−08 (2022) (internal footnotes omitted).

16.   The geofence polygon also embraced part of the interior of the church, and the back yard to the house—places the police would have no reason to believe the suspects had gone. But there is no reason to believe uninvolved persons would have been found in those two places either, at least not between 2:45 and 3:10 o'clock in the morning—even, as here, on a Sunday morning. Moreover, anyone who may, for whatever reason, have been present at those locations at that time would likely at least have heard gunshots, and therefore would at least have been potential witnesses to the offense.

17.   We also granted a second ground for review in this case: “[w]hether the Court of Appeals correctly determined the reliability of Google data[.]” But we now dismiss that ground as improvidently granted.

1.   Both cases upheld multi-day searches of CSLI but were likely abrogated by this Court's decision in Holder because both were decided pre-Carpenter.

2.   Assuming, arguendo, that the warrant in this case is invalid, the good faith exception applies here as well. Here, law enforcement was using cutting-edge technology that, prior to this investigation, was unknown to them. Further, at the time law enforcement obtained and executed the warrant, there was no authority—much less any published authority—on geofence warrants. And law enforcement acted reasonably and with prudence: They wrote the warrant as narrowly tailored, both geographically and temporally, as they could, and they sought an additional warrant for the personal identifying information after the first anonymized list was provided by Google.

3.   See State's Pretrial Exh. 7, at 17 (“Advertising: Google processes information, including online identifiers and information about your interactions with advertisements, to provide advertising. This keeps Google's services and many of the websites and services you use free of charge. You can control what information we use to show you ads by visiting your ad settings.”); see also Chatrie, 107 F.4th at 322 (“But Google uses and benefits from a user opting in, too—mostly in the form of advertising revenue. Google uses Location History to show businesses whether people who viewed an advertisement visited their stores. It similarly allows businesses to send targeted advertisements to people in their stores’ proximity.”)

1.   The warrant ordered that, “[f]or those accounts identified as relevant to the ongoing investigation through an analysis of provided records, and upon demand, the provider shall provide the subscriber's information for those relevant accounts to include, subscriber's name, emails addresses, services subscribed to, last 6 months of IP history, SMS account number, and registration IP.”

2.   Carpenter v. United States, 585 U.S. 296, 303, 138 S.Ct. 2206, 201 L.Ed.2d 507 (2018).

3.   Id. at 296, 138 S.Ct. 2206 (quoting Smith v. Maryland, 442 U.S. 735, 740, 99 S.Ct. 2577, 61 L.Ed.2d 220 (1979)).

4.   Sims v. State, 569 S.W.3d 634, 643 (Tex. Crim. App. 2019) (citing United States v. Jacobsen, 466 U.S. 109, 113, 104 S.Ct. 1652, 80 L.Ed.2d 85 (1984)); King v. State, 670 S.W.3d 653, 656 (Tex. Crim. App. 2023) (“Absent a legitimate expectation of privacy, a defendant lacks standing to raise [a challenge the constitutionality of a search] and we may not consider the substance of his complaint.”).

5.   Wells v. State, 675 S.W.3d 814, 827 (Tex. App. – Dallas 2023, pet. granted) (“Because we conclude that the warrant at issue satisfies the requirements of the Fourth Amendment and, alternatively, Detective Leob's reliance on the warrant was objectively reasonably, it is unnecessary for us to address the State's argument that appellant had no reasonable expectation of privacy in his location history.”); but see King, 670 S.W.3d at 656.

6.   United States v. Knotts, 460 U.S. 276, 281-82, 103 S.Ct. 1081, 75 L.Ed.2d 55 (1983) (“A person traveling in an automobile on public thoroughfares has no reasonable expectation of privacy in his movements from one place to another”). But in United States v. Jones, the Court held that the government's use of a GPS tracking device it installed on the vehicle to monitor a vehicle's movements constituted a search under the Fourth Amendment deciding the case on the basis of the government's physical intrusion into the vehicle. United States v. Jones, 565 U.S. 400, 404-05, 132 S.Ct. 945, 181 L.Ed.2d 911 (2012). The Court acknowledged that “[i]t may be that achieving the same result [as traditional surveillance for a four-week period] through electronic means, without an accompanying trespass, is an unconstitutional invasion of privacy, but the present question does not require us to answer that question.” Id. at 412, 132 S.Ct. 945.

7.   United States v. Miller, 425 U.S. 435, 437, 96 S.Ct. 1619, 48 L.Ed.2d 71 (1976).

8.   Jones, 565 U.S. at 426, 132 S.Ct. 945 (Alito, J., concurring in judgment) (“the use of longer term GPS monitoring in investigations of most offenses impinges on expectations of privacy”); Carpenter, 585 U.S. at 310 n. 3, 138 S.Ct. 2206 (holding that accessing seven days of CSLI constitutes a Fourth Amendment search); see also Sims, 569 S.W.3d at 645-46 (whether government action constitutes a ‘search’ turns on whether ‘enough’ information was seized that it violated a legitimate expectation of privacy); Ford v. State, 477 S.W.3d 321, 335 (Tex. Crim. App. 2015) (holding no expectation of privacy in four days of location data and recognizing that the aggregation of data might be covered by a reasonable expectation of privacy even if a discrete bit of data would not be).

9.   Carpenter, 585 U.S. at 316, 138 S.Ct. 2206. The two orders at issue in Carpenter sought 152 days of cell-site records, which produced records spanning 127 days and seven days of CSLI from a second carrier, which produced two days of records. Id. at 302, 138 S.Ct. 2206. But the Court noted that it was sufficient for their purposes to hold that accessing seven days of CSLI was a search for Fourth Amendment purposes. Id. at 310 n. 3, 138 S.Ct. 2206.

10.   Id. at 311, 138 S.Ct. 2206 (internal citations omitted).

11.   Id. at 309, 138 S.Ct. 2206.

12.   Sims, 569 S.W.3d at 645-46 (“There is no bright-line rule for determining how long police must track a person's cell phone in real time before it violates a person's legitimate expectation of privacy in those records. Whether a person has a recognized expectation of privacy in real-time CSLI records must be decided on a case-by-case basis.”).

13.   Id. at 645.

14.   Id. at 646; see also Holder v. State, 595 S.W.3d 691, 704 (Tex. Crim. App. 2020) (holding, under the Texas Constitution, that there is a reasonable expectation of privacy in 23 days of historical CSLI accessed by the State without probable cause); Ford, 477 S.W.3d at 334-35 (pre-Carpenter holding that obtaining four days of CSLI information did not violate the Fourth Amendment because there was no legitimate expectation of privacy).

15.   United States v. Smith, 110 F.4th 817, 822 (5th Cir. 2024) (citing Haley Amster & Brett Diehl, Note, Against Geofences, 74 Stan. L. Rev. 385, 388 (2022)). “Unlike a warrant authorizing surveillance of a known suspect, geofencing is a technique law enforcement has increasingly utilized when the crime location is known but the identities of the suspects [are] not.” United States v. Rhine, 652 F.Supp.3d 38, 66 (D.D.C. 2023).

16.   Smith, 110 F.4th at 822 (citing Geofence Warrants and the Fourth Amendment, 134 Harv. L. Rev. 2508, 2509 (2021)).

17.   Wells, 675 S.W.3d at 822.

18.   Id.

19.   The suspects had in fact already been captured in the church's parking lot on surveillance video. Id. at 823.

20.   Id. at 822.

21.   Sims, 569 S.W.3d at 646 (no legitimate expectation of privacy in physical movements or location as reflected in less than three hours of real-time CSLI); see also United States v. Chatrie, 107 F.4th 319, 330 (4th Cir. 2024) (holding there was no reasonable expectation of privacy in two hour’ worth of location history data obtained from Google).

22.   Carpenter, 585 U.S. at 311, 138 S.Ct. 2206 (noting that 127 days of time-stamped data provides an intimate window into a person's life, revealing not only his particular movements, but through them his familiar, political, professional, religious, and sexual associations and that these location records thus hold the privacies of life).

23.   Id. at 315, 138 S.Ct. 2206.

24.   Id. (citing Smith, 442 U.S. at 745, 99 S.Ct. 2577).

25.   See id. (noting that “[a]part from disconnecting the phone from the network, there is no way to avoid leaving behind a trail of location data”). Here, on the other hand, at the Fourth Circuit has recognized, “[w]hether Google tracks a user's location ․ is entirely up to the user himself. If Google compiles a record of his whereabouts, it is only because he has authorized Google to do so.” Chatrie, 107 F.4th at 331.

26.   Smith, 442 U.S. at 743-44, 99 S.Ct. 2577.

27.   Carpenter, 585 U.S. at 314-15, 138 S.Ct. 2206.

28.   Id. at 319, 138 S.Ct. 2206.

29.   State v. Granville, 423 S.W.3d 399, 405-06 (Tex. Crim. App. 2014).

30.   Carpenter, 585 U.S. at 311, 138 S.Ct. 2206.

31.   Id.

32.   State v. Duarte, 389 S.W.3d 349, 354 (Tex. Crim. App. 2012).

33.   State v. Baldwin, 664 S.W.3d 122, 123 (Tex. Crim. App. 2022). Although the geofence warrant at issue involved a search of Google's location history database, ultimately the detectives used the information gleaned to search Appellant's phone without providing any nexus between the phone and the offense beyond boilerplate assumptions about the use of phones in home invasion offenses.

34.   Id. at 132 (citing Rodriguez v. State, 232 S.W.3d 55, 61 (Tex. Crim. App. 2007)); Duarte, 389 S.W.3d at 354.

35.   Baldwin, 664 S.W.3d at 134.

36.   Id. at 125.

37.   Id. at 126.

38.   Id.

39.   Id. at 135 (“The boilerplate language in itself is not sufficient to provide probable cause in this case, nor does the remaining affidavit set forth details in sufficient facts to support probable cause. Considering the whole of the affidavit, there is no information included that suggest anything beyond mere speculation that [the defendant's] cellphone was used before, during, or after the crime.”).

40.   Id. (citing Tolentino v. State, 638 S.W.2d 499, 502 (Tex. Crim. App. 1982)).

41.   Id. at 134.

42.   Id. at 123 (While boilerplate language may be used in a search warrant affidavit, to support probable cause, “the language must be coupled with other facts and reasonable inferences that establish a nexus between the device and the offense.”).

43.   Certainly, no probable cause was established to support a search of the cell phones or the data held by the cell phones, which the warrant sought when it requested historical IP data. There is a real danger in concluding otherwise that a lowered standard for probable cause will be used to search cell phones and their data with the attendant privacies of life themselves. In fact, in this case, secondary warrants to search the phone accounts identified through this geofence warrant were obtained based upon little more information than that used to justify the geofence warrant.

44.   Duarte, 389 S.W.3d at 354 (quoting Illinois v. Gates, 462 U.S. 213, 239, 103 S.Ct. 2317, 76 L.Ed.2d 527 (1983)).

45.   Indeed, the surveillance video in this case does not provide any factual support for the “keep an open line” conclusion contained in the warrant affidavit.

46.   Smith, 110 F.4th at 836 (“[T]he Fourth Amendment was the founding generation's response to the reviled ‘general warrants’ and ‘writs of assistance’ of the colonial era, which allowed British officers to rummage through homes in an unrestrained search for evidence of criminal activity. ‘General warrants’ are warrants that ‘specif[y] only an offense,’ leaving ‘to the discretion of the executing officials the decision as to which persons should be arrested and which places should be searched.’ ”) (internal citations omitted).

47.   Estrada v. State, 154 S.W.3d 604, 609 (Tex. Crim. App. 2005) (“Probable cause to search exists when reasonably trustworthy facts and circumstances within the knowledge of the offense on scene would lead a man of reasonable prudence to believe that the instrumentality of a crime or evidence of a crime will be found.”). Surely, locating and identifying a person near a particular crime scene does not provide sufficient probable cause to search their phone. Here, locating and identifying potential witnesses in and of itself is not contraband nor evidence of a crime.

48.   Wells, 675 S.W.3d at 825 (“The first search revealed five devices within the geofence ․ [t]he second stage of the search ․ indicated that two of the devices travelled past the gas station where the suspect's car was recorded on video surveillance. The third stage of the search revealed the identity of the Google account subscribers for the two devices identified as relevant.”).

49.   Carpenter, 585 U.S. at 311, 138 S.Ct. 2206.

50.   Riley v. California, 573 U.S. 373, 395, 134 S.Ct. 2473, 189 L.Ed.2d 430 (2014).

51.   Carpenter, 585 U.S. at 311-15, 138 S.Ct. 2206 (acknowledging the fact that people compulsively carry cell phones in support of a greater privacy concern for historical cell-site records and that cell phones are a pervasive and insistent part of daily life as a reason not to apply the third-party doctrine to such records); Riley, 573 U.S. at 395, 134 S.Ct. 2473 (recognizing the pervasive character of cell phones, which “carry a cache of sensitive personal information,” as support for not dispensing with the warrant requirement).

Yeary, J., announced the judgment of the Court and filed an opinion in which Keel, Finley, and Parker, JJ., joined.

Finley, J., filed a concurring opinion in which Parker, J., joined. Newell, J., filed a concurring and dissenting opinion in which Richardson and Walker, JJ., joined. McClure, J., dissented. Schenck, P.J., did not participate.