UNITED STATES DISTRICT COURT FOR THE EASTERN DISTRICT OF VIRGINIA Norfolk Division LEE SCHMIDT and CRYSTAL ) ARRINGTON, ) ) Plaintiffs, ) ) v. ) Case No. 2:24cv621 ) CITY OF NORFOLK, the NORFOLK ) POLICE DEPARTMENT, and MARK ) TALBOT, in his official capacity ) as the Norfolk Chief of Police, ) ) Defendants. )
MEMORANDUM OPINION AND ORDER Defendants the City of Norfolk! and Norfolk Chief of Police Mark Talbot’s (collectively, “Defendants”) Motion to Compel, ECF No. 81, is before the Court. Plaintiffs Lee Schmidt and Crystal Arrington (collectively, “Plaintiffs”) sued Defendants on October 21, 2024. ECF No. 1. Plaintiffs allege that by operating and accessing an automatic license plate reader (“ALPR”) system supplied by private contractor Flock Safety (“Flock”)’, Defendants are violating their Fourth Amendment right to be free from unreasonable searches. /d. at §] 93-109; ECF No. 29 at 4, Defendants filed the Motion to Compel and an accompanying memorandum in support on June 9, 2025. ECF No. 81. Plaintiffs filed an Opposition to the Motion to Compel (“Opposition”)
' On December 13, 2024, the Court entered a consent order dismissing the Norfolk Police Department from the case without prejudice. ECF No. 21. ? The Court refers to the ALPR system supplied by Flock, including its centralized database and machine learning capabilities as described by Plaintiffs, see ECF No. | at §] 2-3, as “the Flock system” throughout this opinion. The Court uses the term “Flock cameras” to refer to the literal cameras that form the base of this system. See id.
on June 23, 2025. ECF No. 88. Defendants filed a Reply to the Opposition (“Reply”) on June 30, 2025, ECF No. 94, and the Court held a hearing on the Motion to Compel on August 6, 2025. Thus, the Motion to Compel is now ripe for disposition. For the reasons explained below, Defendant’s Motion to Compel is DENIED. I, FACTUAL AND PROCEDURAL BACKGROUND According to the Complaint, Defendants operate a network of 172 Flock cameras placed throughout Norfolk. ECF No. 1 at ¢ 2. The Flock cameras photograph cars that pass them. Jd. Then, using the images captured by the cameras, the Flock system analyzes each car’s license plate, color, make, model, and unique identifying features (like bumper stickers) to make a “vehicle fingerprint” for that car. Jd. at 33. The system maintains each vehicle fingerprint in a centralized database. Jd? By accessing that database and the vehicle fingerprints, a Flock system user can “link[] together different photos of the same car and creat[e] a detailed record of [each car’s] movements.” /d. at | 41. Plaintiff Lee Schmidt lives in Norfolk and near several Flock cameras. /d. at J] 51, 53. Plaintiff Crystal Arrington operates her own home healthcare business and frequently visits Norfolk to care for her clients and accompany them to appointments, and she also visits the city to see family and friends. Jd at § 62-63. Both Plaintiffs allege that the Flock system allows Defendants to “create a running 30-day record of the whole of [their] movements throughout Norfolk ....” Jd. at] 74. They argue that by creating this record, Defendants are violating their Fourth Amendment right to be free of unreasonable searches because they have a reasonable expectation of privacy in the “whole of their long-term physical movements.” Jd. at { 99a.
3 Flock usually maintains the data it collects for thirty days, according to the Complaint. ECF No. 1 at ] 43. But the system may store the data for a longer period of time in specific circumstances. Id.
Defendants moved to dismiss the Complaint pursuant to Federal Rules of Civil Procedure 12(b)(1) and 12(b)(6), arguing that the Court lacked subject-matter jurisdiction to decide Plaintiffs’ claims and that Plaintiffs ultimately failed to state a claim. ECF No. 18. The Court denied the Motion to Dismiss, holding that it had subject-matter jurisdiction to consider Plaintiffs’ claims and that Plaintiffs, in claiming that Defendants were using the Flock system to create a long-term catalogue of their movements, “plausibly alleged that a warrantless search occurred, and thus interests that the Fourth Amendment was designed to protect have been violated.” ECF No. 29 at 20. The Court noted that discovery would help clarify several important and unresolved factual issues, including “the location of the Flock cameras throughout the city . . . the similarities, or lack thereof, between the Flock camera distribution in this case and the Flock camera distribution in Richmond at issue in United States v. Martin, [753 F. Supp. 3d 454, 461 (E.D. Va. 2024)],” and the “nature of the Flock database and its use or misuse by [the Norfolk Police Department], the extent to which Plaintiffs’ vehicles have actually been captured in the Flock system, and Defendants’ ability to effectively “track” Plaintiffs’ vehicles as they move through Norfolk.” Jd. at 19 n.9, 20 n.10. The case progressed and discovery began. The Motion to Compel relates to seven interrogatories and six requests for production (“RFP”) Defendants issued to Plaintiffs. ECF No. 82 at 19-22. These requests seek two types of information. ECF No. 82 at 12. First, they seek documentation of Plaintiffs’ physical movements. Id.4 So, for example, Interrogatory Thirteen requests that Plaintiffs provide their “location throughout the dates for which [Plaintiffs] have requested that Defendants provide data, including the address and individuals present with you.” ECF No. 83, attach. 2, at 6. Second, they seek
The following disputed RFPs fit into this category: four, five, nine, twelve, fifteen, and sixteen. ECF No. 82 at 19-22. The following disputed interrogatories fit into this category: three, seven, and thirteen. Jd.
information about the people with whom Plaintiffs most often associate. ECF No. 82 at 12.5 An example of this type of request is Interrogatory Five: “[i]dentify the ten individuals that YOU have interacted with most frequently in the city of Norfolk in the REPORTING PERIOD and the date on which YOU interacted with those individuals.” ECF No. 83, attach. 2 at 5. Plaintiffs objected to the disputed interrogatories and RFPs as irrelevant, overbroad, unduly burdensome, and in some cases, seeking information protected by the First Amendment. ECF No. 88, attach. 1 at 1-19.° At the hearing on the Motion to Compel, for the reasons stated on the record at that hearing, the Court denied the Motion to Compel as to every request for production and interrogatory except RFP Sixteen, which is now the only discovery request that remains in dispute.’ RFP Sixteen asks Plaintiffs to provide “all Cell Site Location Information (“CSLI”) between February 19, 2025 and the present.” ECF No. 83, attach. 14 at 5. At the hearing, Defendants said that they believed that the CSLI data was the most practical and relevant information to their defenses and Plaintiffs’ claims. The Court took the issue of whether Plaintiffs should be compelled to produce their CSLI records under advisement.”
>No disputed RFPs fit into this category, but the following disputed interrogatories, issued to the Plaintiffs individually and crafted to fit the specific allegations each Plaintiff set forth in the Complaint, do: Schmidt Interrogatory Five, Schmidt Interrogatory Fourteen, ECF No. 83, attach. 4 at 5, 7, Arrington Interrogatory Fifteen, and Arrington Interrogatory Sixteen, ECF No. 83, attach. 2 at 7. § Plaintiffs produced some information in response to the interrogatories and requests for production consistent with their objections. See ECF No. 88, attach. | at 1-19 ? The Court held that the other discovery requests were either unduly burdensome to produce or irrelevant to Defendants’ defenses. See infra note 17. 8 Defendants served RFP Sixteen on Plaintiffs for the first time on the same day they moved to compel Plaintiffs’ responses to the other disputed discovery requests. Although Plaintiffs had not specifically objected to RFP Sixteen when they filed their Opposition—as the time to do so had not expired—they argued against the production of CSLI in the Opposition and at the hearing on the Motion to Compel. See ECF No. 88 at 26-30. On August 4, 2025, two days before the hearing on the Motion to Compel, Plaintiffs filed a Notice informing the Court that Defendants “issued three subpoenas demanding CSLI from Plaintiffs’ cellphone
The Court will first address whether Defendants’ requests for that information is relevant and proportional to the needs of the case pursuant to Federal Rule of Civil Procedure 26. II. LEGAL STANDARDS A. The scope of discovery Federal Rule of Civil Procedure 26(b)(1) allows parties to “obtain discovery regarding any nonprivileged matter that is relevant to any party’s claim or defense and proportional to the needs of the case....” The “threshold for relevance is not high one... □□ Cont. Materials Processing, Inc. v. Kataleuna GmbH Catalysts, 462 F. App’x 266, 273 (4th Cir. 2012). Indeed, the term “relevant” in the context of Rule 26 “has been ‘broadly construed to encompass any possibility that the information sought may be relevant to the claim or defense of any party.’” Pecos River Talc LLC v. Emory, No. 4:24CV75, 2025 WL 1888565, at *3 (E.D. Va. July 8, 2025) (quoting EEOC v. Sheffield Fin., LLC, No. 1:06cv889, 2007 WL 1726560, at *3 (M.D.N.C. June 13, 2007). The standard for relevance in discovery is broader than the standard for relevance set forth by the Federal Rules of Evidence. /d. (citing Tucker v. Momentive Performance Materials USA, Inc., No. 2:13cv4480, 2016 WL 8252929, at *3 (S.D.W. Va. Nov. 23, 2016)). In determining whether evidence is relevant and proportional, Rule 26 directs the Court to consider the following: “the importance of the issues at stake in the action, the amount in controversy, the parties’ relative access to relevant information, the parties’ resources, the importance of the discovery in resolving the issues, and whether the burden or expense of the proposed discovery outweighs its likely benefit.” Fed. R. Civ. P. 26(b)(1); Walsh v. Kynd Hearts Home Healthcare, LLC, No. 2:20CV630, 2022 WL 18108388, at *1 (E.D. Va. Sept. 1, 2022). service providers.” ECF No. 97 at 1. According to the Notice, Plaintiff Schmidt’s cellphone service provider complied with one of the subpoenas’ requests and produced Plaintiff Schmidt’s CSLI data to Defendants. Jd. The Notice and the incident it describes have no bearing on the Court’s present inquiry as to the Motion to Compel, but is addressed infra Section III.B.3.
B. Moving to compel a discovery response When one party fails to respond to discovery requests, the requesting party may move to compel a discovery response pursuant to Federal Rule of Civil Procedure 37(a)(3)(B). Rule 37(a)(3)(B)(iii) permits a party to file a motion to compel an answer to an interrogatory submitted under Federal Rule of Civil Procedure 33, while Rule 37(a)(3)(B)(iv) allows a party to file a motion to compel the opposing party to produce documents. “On a motion to compel, ‘the burden of proof is with the party objecting to the discovery to establish that the challenged production should not be permitted.”” Doe v. Old Dominion University, 289 F. Supp. 3d 744, 749 (E.D. Va. 2018) (quoting Singletary v. Sterling Transp. Co., 289 F.R.D. 237, 241 (E.D. Va. 2012)). C. The Fourth Amendment and drag-net surveillance systems Whether Plaintiffs have carried their burden to prove that Defendants’ request for their CSLI data is not relevant or proportional to the needs of the case!® depends on what Plaintiffs must ultimately prove to establish a violation of their Fourth Amendment rights. See Fed. R. Civ. P. 26(b)(1) (directing the Court to consider not only relevance but also “the importance of the discovery in resolving the issues” and “whether the burden or expense of the proposed discovery outweighs its likely benefit.”); Ramos v. Walmart, Inc., No. CV 21-13827, 2023 WL 2326965, at *4 (D.N.J. Mar. 2, 2023) (“The Rule 26 questions of relevance and proportionality, in turn, must be informed by the claims and issues presented in the action in which discovery is sought.”). Thus, before the Court addresses the parties’ respective arguments relating to the production of CSLI data, a discussion of the relevant Fourth Amendment standard at issue is necessary.
Although, as explained below, Plaintiffs argue that the Court cannot order them to consent to the production of their CSLI, it is not clear whether they also argue that their CSLI is privileged. The Court holds that Plaintiffs have shown that the request is not proportional to the needs of the case, so it need not rule on Plaintiffs’ potential privilege argument.
The Fourth Amendment states that the “right of the people to be secure in their persons, houses, papers, and effects, against unreasonable searches and seizures, shall not be violated . . .” U.S. Const. amend. IV. In its Memorandum Opinion Denying Defendants’ Motion to Dismiss, ECF No. 29, the Court recognized that federal courts apply either the “common-law property- based approach”! or the “Karz test” in determining whether a “search” has occurred. ECF No. 29 at 9-10; Katz v. United States, 389 U.S. 347, 361 (1967) (Harlan, J., Concurring). In denying Defendants’ Motion to Dismiss, the Court applied the Katz test. ECF No. 29 at 10. The Katz test asks two questions to determine whether a search has occurred. Katz, 389 USS. at 361. First: does the individual have a subjective expectation of privacy in the area or thing allegedly being searched? Jd. Second: is the individual’s belief that the area or thing is private one that society recognizes as reasonable, or in other words, is there an objective expectation of privacy in that area or thing? Jd. Ifthe answer to either question is no, no search has occurred; if the answer to both questions is yes, a search has occurred. ECF No. 29 at 10 (citing United States v. Martin, 753 F. Supp. 3d 454, 461 (E.D. Va. 2024)). Thus, to prove that an unreasonable search has occurred in violation of the Fourth Amendment, a plaintiff must show that the challenged activity or surveillance invaded an area or thing in which the plaintiff maintained both an objective and subjective expectation of privacy. Katz, 389 U.S. at 361. Federal courts apply the Kafz test to determine whether a search occurs when the government uses large-scale surveillance systems to monitor an individual’s movements. In Carpenter v. United States, the Supreme Court recognized that individuals maintain both a
'! The Court recognized that the Karz test “has recently been reaffirmed as an alternative to the property- based approach in various Supreme Court holdings” and did not apply the “common-law property-based approach” in analyzing Plaintiffs’ claims. ECF No. 29 at 10 n.5 (citing Carpenter v. United States, 585 U.S. 296, 304 (2018); United States v. Jones, 565 U.S. 400, 406 (2012); Kyllo v. United States, 533 U.S. 27, 33 (2001)).
subjective and objective expectation of privacy in the “whole of their physical movements.” 585 U.S. 296, 310 (2018); ECF No. 29 at 11-12 (discussing Carpenter). Therefore, when the government!? accessed, without a warrant, periods of a criminal defendant’s CSLI data that “placed [him] within a wedge-shaped sector ranging from one-eighth to four square miles” and allowed the government to “in combination with other information, deduce a detailed log of [his] movements . . . ,” it conducted an unreasonable search in violation of the Fourth Amendment. Id. at 302, 312. The Court emphasized that the “retrospective quality” of the CSLI data and the fact that individuals “compulsively carry cell phones with them all the time” meant that the CSLI data, when combined with other information garnered through more traditional investigative techniques, gave the Government a clear picture of the defendant’s movements over time. Jd. at 311-313. And because a detailed log of all physical movements provides “an intimate window into a person’s life,” the Government contravened the defendant’s subjective and objective expectation of privacy in the whole of his physical movements when it accessed the CSLI data. /d. at 311. Sitting en banc, the Fourth Circuit applied Carpenter in a different context in Leaders ofa Beautiful Struggle v. Baltimore Police Department. 2 F Ath 330, 340 (4th Cir. 2021). In that case, the Fourth Circuit examined the Aerial Investigation Research (“AIR”) surveillance system that featured multiple planes flying above and photographing the city of Baltimore, Maryland for extended periods of time. Jd. at 333-34. The planes flew during the daylight hours—not at night or during bad weather—and provided continuous photographic coverage of approximately ninety percent of the city for a total of about forty hours every week. Jd. The images the planes captured could be “magnified to a point where people and cars are individually visible, but only as blurred dots or blobs” and were stored on servers for forty-five days. Jd. at 341. Even though the AIR
'2 The Court uses the term “the government” broadly in this opinion to refer to the entity that conducts the alleged search.
system by itself did not provide a clear, identity-revealing picture, the Court recognized that the government could “deduce an individual’s identity from AIR data, other available information, and some deductive reasoning,” and that AIR data combined with other surveillance information enabled the government to “glean insights” or enable “deductions from the whole of individuals’ movements.” /d. at 344-45, The Court held that accessing the data gathered by the AIR system violated the Fourth Amendment. /d. at 346. The Court recognized that the AIR system did not allow “perfect tracking of all individuals it captures across all the time it covers,” but because it enabled “photographic, retrospective location tracking in multi-hour blocks, often over consecutive days, with a month and half of daytimes for analysts to work with,” it provided a “‘wealth of detail,” greater than the sum of the individual trips.” Jd. at 342 (citing United States v. Jones, 565 U.S. 400, 415-17 (2012) (Sotomayor, J., concurring)). Because of this capability, the AIR system conducted surveillance closer to the long-term tracking of the totality of an individuals’ movements present in Carpenter than the short-term surveillance of which police are traditionally capable.!? Jd. at 341. Thus, the Court held that “[blecause the AIR program enables police to deduce from the whole of
The Fourth Circuit recognized that “[t]he touchstone in Carpenter was the line of cases addressing ‘a person’s expectation of privacy in [their] physical location and movements.’” Leaders of a Beautiful Struggle, 2 F.4th at 340 (quoting Carpenter, 585 U.S. at 306). One case in that line is United States v. Knotts, where the Supreme Court held that placing a tracking device inside a drum of chloroform and then using it to follow the car in which the drum was placed did not constitute a search. 460 U.S. 276, 285 (1983). The Court reasoned that the device merely allowed the government to easily surveil the car as it travelled on public roads, as it could have done by assigning a police officer to tail the car. Jd. “Nothing in the Fourth Amendment prohibited the police from augmenting the sensory faculties bestowed upon them at birth with such enhancement as science and technology afforded them in this case.” /d. at 282. Another case in the line is United States v. Jones. 565 U.S. 400, 413 (2012). There, the government placed a GPS tracking device on an individual’s car and used it to record the vehicle’s movements over a period of twenty- eight days. Jd. at 403. The Court decided that case by holding that the government violated the Fourth Amendment by trespassing on the defendant’s personal property, id. at 410, but five justices recognized that long-term GPS monitoring would infringe on an individual’s reasonable expectation of privacy in the whole of their vehicle’s movements. Jd. at 415 (Sotomayor, J., concurring); id. at 430 (Alito, J., concurring in the judgment).
individuals’ movements, we hold that accessing its data is a search, and its warrantless operation violates the Fourth Amendment.” Jd. at 346. Even with substantial gaps in surveillance at night or during bad weather, the AIR system captured enough of the Baltimore residents’ daily movements to qualify as a search pursuant to Carpenter. Id. at 342, 346. Crucially, Leaders of a Beautiful Struggle demonstrated that Carpenter’s Fourth- Amendment rationale could apply to a surveillance system with substantial gaps in coverage. Id. Carpenter recognized that imprecisions in the CSLI technology did not detract from the overall logging of the “whole of an individuals’ physical movements” because “inference” does not “insulate a search.” 585 U.S. at 312 (quoting Kyllo v. United States, 533 U.S. 27, 36 (2001)). Leaders of a Beautiful Struggle held that daily twelve-hour gaps in surveillance did not defeat the AIR system’s ability to violate individuals’ reasonable expectation of privacy in the whole of their movements. 2 F.4th at 346. Although it recognized these gaps, the Fourth Circuit did not decide the case by providing a bright-line rule as to how much of a person’s movements a surveillance system must capture to become a drag-net system in violation of Carpenter. See id. at 341 (“Thus, Carpenter solidified the line between short-term tracking of public movements—akin to what law enforcement could do ‘prior to the digital age’-—and prolonged tracking that can reveal intimate details through habits and patterns.”) (quoting Carpenter, 585 U.S. at 310).'* Instead, it “hinged its analysis on Carpenter’s solidification of the line ‘between short-term tracking of public movements . . . and prolonged tracking that can reveal intimate details through habits and
'4 In United States v. Sturdivant, the Court recognized the absence of this bright-line rule: “[L]ower court acceptance of ALPR databases leaves serious doubt about the point, if any, at which governmental use of cameras crosses the line to an impermissible warrantless search . . . Such surveillance could become too intrusive and run afoul of Carpenter at some point. But when?” United States v. Sturdivant, No. 1:22-CR- 615, 2025 WL 1633754, at *9 (N.D. Ohio June 9, 2025). 10
patterns.” Martin, 753 F. Supp. 3d at 465 (quoting Leaders of a Beautiful Struggle, 2 F.4th at 341). Plaintiffs’ case is not the first to raise the issue of whether ALPR surveillance qualifies as a drag-net system that invades a reasonable expectation of privacy pursuant to Carpenter and Leaders of a Beautiful Struggle. See United States v. Sturdivant, No. 1:22-CR-615, 2025 WL 1633754, at *8 (N.D. Ohio June 9, 2025) (collecting cases examining challenges to the use of ALPR databases on Fourth Amendment grounds).'° This Court addressed that issue in the criminal context in United States v. Martin, where the defendant challenged the government’s use of Flock cameras and their accompanying database to gather evidence against him. 753 F. Supp. 3d at 461. Strictly relying on the facts before it, the Court held that the Flock system only captured “three individual snapshots of [the defendant’s] brief location at specific times,” so using the system and accessing its data “hardly [rose] to the level of persistent, unceasing public surveillance that the courts found troublesome in Carpenter or Beautiful Struggle.” Id. at 473.'®
15 Neither the Court nor the parties found a civil case challenging the use of ALPR systems pursuant to the Fourth Amendment that progressed beyond the motion to dismiss stage and into discovery. In Scholl v. Illinois State Police et al., the plaintiffs sued defendants pursuant to 42 U.S.C. § 1983 and challenged Illinois’s warrantless use of ALPRs on Fourth Amendment grounds. 776 F. Supp. 3d 701, 707 (N.D. Ill. 2025). However, the Court granted the defendants’ motion to dismiss. Jd. at 721. The plaintiffs filed a notice of appeal of the Court’s order dismissing their complaint in that case in May 2025. See ECF No. 45, Scholl vy. Illinois State Police et al., No. 24-cv-4435 (N.D. II. May 15, 2025). '6 Martin distinguished the Flock system from the aerial surveillance at issue in Leaders of a Beautiful Struggle by emphasizing that in Baltimore, through the use of the aerial surveillance system, the government could track and monitor “every Baltimore resident’s movements during the daylight hours, every day,” and that this level of movement-tracking allowed police an in-depth look on an individual’s activities and associations. 753 F. Supp. 3d at 471. Because the Flock system only captured three images of the car, the surveillance was not nearly as pervasive as the aerial system in Leaders of a Beautiful Struggle. Id. at 473. Relying on Knotts, the Court held that the Flock system did not surveille the drivers of the car it photographed, it merely “augment[ed] the same inherent sensory faculties of law enforcement that have existed since the founding.” Jd. at 476 (citing Knotts, 460 U.S. at 282).
In United States v. Sturdivant, the Court denied a criminal defendant’s motion to suppress evidence obtained from two separate ALPR systems. 2025 WL 1633754, at *11. The Court ultimately denied the motion to suppress because the Defendant had no reasonable expectation of privacy “in the appearance of his vehicle or in his license plate number,” so accessing that information through the ALPR databases did not constitute a search. Jd at *6. But the Court offered a detailed explanation of why it found that Carpenter and Beautiful Struggle did not compel a different result, and like the court in Martin, relied heavily on the specific facts of the case. Id at *7. The ALPR surveillance systems in that case gathered twenty-six images of the defendant’s car over a period of six weeks from an unknown number of search results. Jd. at *11. The Court held that in providing those images, the databases merely “provided discrete data points with considerable stretches of obscurity in between. It is true that inference cannot insulate a search from Fourth Amendment scrutiny . . . [bJut no amount of inference could create a ‘detailed log of [the defendant’s] movements’ from the data at issue here.” /d. at *10 (citing Carpenter, 585 U.S. at 12) (internal citations omitted). The Court did note that, because of the number of images captured by the ALPR systems at issue, “compared to most other cases involving ALPR, the record shows a qualitative leap forward, placing the evidence generated from [ALPR database queries] at risk in the absence of a warrant.” Jd. at *11 (comparing the number of images captured by the ALPR system in the case before the Court to the number of images captured by similar systems in other cases). Ill. ANALYSIS Plaintiffs and Defendants dispute whether, given the framework described above, Plaintiffs’ CSLI data is relevant and proportional to Defendants’ defense against Plaintiffs’ Fourth Amendment claims. The Court addresses each Rule 26 requirement individually.
A. Plaintiffs’ CSLI data has some relevance to their claims and Defendants’ defenses. Defendants argue that the same relevance arguments apply to each of their discovery requests for information about Plaintiffs’ movements.'? They claim that they need information about the Plaintiffs’ whereabouts to “test” the Plaintiffs’ claims that “Flock camera data can track whole of [their] movements,” and that they cannot conduct this test “without knowing Plaintiffs’ location throughout the relevant period, and with whom they associate.” ECF No. 82 at 13. They argue that “the central issue in this case . . . is whether or not Flock camera data can track the whole of Plaintiffs’ movements and reveal intimate and private details of their lives, including where they go and with whom they associate.” Jd. at 12. Without CSLI data, Defendants claim that neither they nor the Court will know how many of the Plaintiffs’ movements the Flock system captures, will see “only those times Plaintiffs visited a location after passing a nearby Flock camera...,” and will be unable to corroborate the Plaintiffs’ recounting of their whereabouts. Id. at 13. Plaintiffs counter that “the relevant data in this case is what Defendants have collected, not information they missed.” ECF No. 88 at 15. Plaintiffs emphasize that the test for whether use of the Flock system qualifies as a drag-net surveillance operation depends not on “whether [the Flock system] captures literally 100 percent of people’s movements,” but instead whether its use is more
'7 In this Memorandum Opinion, the Court only considers the parties’ relevance and proportionality arguments with respect to Defendants’ requests for CSLI data, even though those arguments applied to Defendants’ request for other information as well. See ECF No. 88 at 27 n. 14 (citing the general relevance arguments raised earlier in Plaintiffs’ Opposition); ECF No. 94 at 6~9 (providing a general relevance argument related to Defendants’ requests for location information). This is because, at the hearing, the Court denied the Motion to Compel with respect to every discovery request apart from RFP Sixteen, which requested Plaintiffs’ CSLI data, based on its affirmative finding that the type of information sought by the other requests, for example, financial records, odometer readings, and the ten people with whom Plaintiffs interact most frequently, were decidedly not relevant. ECF No. 83, attach. 14 at 5; see ECF No. 83, attach. 2 at 5; ECF No. 83, attach. | at 7. 13
like the long-term surveillance at issue in Carpenter and Leaders of a Beautiful Struggle than traditional short-term surveillance. /d. (citing Leaders of a Beautiful Struggle, 2 F Ath at 343, 345), They argue that the existence of gaps in the surveillance captured by the AIR system in Leaders of a Beautiful Struggle prove this point, and that the real question before the Court is what Defendants may deduce about them by combining the Flock data with other information gathered through different surveillance techniques. /d. (citing Leaders of a Beautiful Struggle, 2 F.4th at 344-45). At the hearing, Plaintiffs also proffered that their expert witnesses relied on information already produced or available to Defendants. In their Reply, Defendants argue that Plaintiffs misstate the legal standard that governs their claims by deemphasizing the question of whether a surveillance system tracks the literal entirety of a person’s movements. ECF No. 94 at 7. They also note that they need location information to understand the “gaps” in the Flock system’s surveillance efforts and “draw factual comparisons to other Fourth Amendment cases.” /d. at 8. To determine whether Plaintiffs’ CSLI data is relevant under Federal Rule of Civil Procedure 26, the Court must decide whether there is a “possibility that [Plaintiffs’ CSLI data] may be relevant to the claim or defense of any party.’”” This standard is not difficult to meet, Cont. Materials Processing, Inc., 462 F. App’x at 273, and the Court finds that Defendants have met it. When it denied Defendants’ Motion to Dismiss, the Court emphasized that “the complaint alleges facts notably similar to those in Carpenter that the Supreme Court found to clearly violate society’s expectation of privacy: law enforcement secretly monitoring and cataloguing the whole of tens of thousands of individuals’ movements over an extended period.” ECF No. 29 at 19. And the Court repeatedly emphasized that the Plaintiffs alleged that the Flock system “tracks the whole of Plaintiffs’ physical movements.” /d. at 15; see id. at 19. Moreover, the Supreme Court in
Carpenter and the Fourth Circuit in Leaders of a Beautiful Struggle noted that one of the Fourth Amendment issues posed by those cases was their ability to precisely track every individual’s movements over time. Carpenter, 585 U.S. at 312; Leaders of a Beautiful Struggle, 2 F.4th at 342. As Defendants recognized at the hearing, the Fourth Circuit’s opinion in Leaders of a Beautiful Struggle did compare the capabilities of the AIR technology to the CSLI data reviewed in Carpenter. See Leaders of a Beautiful Struggle, 2 F.4th at 341-43. This suggests that a comparison of Plaintiffs’ recorded movements and the data the Flock system captured could be helpful to Defendants’ efforts to compare the facts of this case to those of Carpenter and others to show that surveillance conducted through the Flock system is less pervasive.'® And in Sturdivant, the Court briefly compared the data collected by the ALPR system at issue in that case to the defendant’s own account of his public movements to determine whether the ALPR system, on its own, could provide the government the same information. Sturdivant, 2025 WL 1633754, at *10. Thus, a detailed record of Plaintiffs’ movements, provided via CSLI data, would be at least partially relevant to the parties’ claims and defenses because it could allow Defendants to compare the Flock system to other surveillance systems and could provide examples of trips the Flock system failed to catch, potentially strengthening the Defendants’ arguments about the gaps in Flock system surveillance. B. Plaintiffs’ CSLI data is not proportional to the needs of the case because it is minimally important to the claims at issue and runs counter to Plaintiffs’ privacy interests. The parties’ proportionality arguments focus primarily on the burden of producing the requested discovery. See ECF No. 82 at 14; ECF No. 88 at 24; ECF No. 94 at 11-12. Although the Court recognizes the burden the request imposes on Plaintiffs’ privacy interests, it also finds
'8 However, as the Court noted in Sturdivant, “[b]ecause of the significant differences between cell-site location information and automatic license plate reader data, direct comparison of [the] quantitative figures [each system provides] is not necessarily helpful.” 2025 WL 1633754, at *10. 15
two other Rule 26 considerations—the importance of the issues at stake in the action and the importance of the discovery in resolving the issues—tip the scale in favor of denying the Motion to Compel. Fed. R. Civ. P. 26(b)(1). There is a superficial appeal to the notion that outside information, separate and apart from the data collected by the Flock system, can be utilized as a check on the conclusions reached by using the Flock system data. In sum, Defendants want to use the CSLI data to show every trip the Plaintiffs took that the Flock system failed to log, so that they can demonstrate that it captures something less than the “whole” of Plaintiffs’ movements. They distinguish this case from Carpenter and Leaders of a Beautiful Struggle by demonstrating that the Flock system simply does not capture the same percentage of the Plaintiffs’ movements that CSLI data, or constant aerial photography, could. But the importance and relevance of this showing to the resolution of the case is limited for three reasons. First, a surveillance system that captures something less than the whole of an individual’s movements may still qualify as a drag-net system under Carpenter. Second, courts determine whether a surveillance system is a drag-net system by analyzing the data that system collects and determining what it shows the government. Third and finally, ordering Plaintiffs to turn over their CSLI data to the government would raise significant privacy concerns. The Court addresses each reason in turn. 1. A surveillance system can capture something less than the whole of an individual's movements and still qualify as a drag-net system under Carpenter. In Leaders of a Beautiful Struggle, the Fourth Circuit recognized that a surveillance system could qualify as an unconstitutional drag-net system under Carpenter even if it did not capture the entirety of individuals’ movements with the precision of CSLI data. 2 F.4th at 346. Daily twelve- hour gaps in surveillance, which meant that the AIR system could not possibly track every individual’s every public movement, did not affect this holding. /d. at 342,344. Thus, Carpenter’s 16
emphasis on CSLI data capturing the “whole of [an individual’s] physical movements,” 585 U.S. at 313, did not mean that a surveillance system that captured something less than the literal entirety of a person’s physical movements could not violate an individual’s reasonable expectation of privacy in the whole of their physical movements. Leaders of a Beautiful Struggle, 2 F Ath at 346. Consequently, CSLI data demonstrating that the Flock system did not capture the literal entirety of the Plaintiffs’ movements would not definitively resolve the Fourth Amendment issue, even though it may have some relevance. Defendants argue that they need Plaintiffs’ CSLI data to contextualize and determine the size of the gaps in the Flock system’s surveillance. ECF No. 94 at 8. But in Leaders ofa Beautiful Struggle, the existence of gaps in the surveillance captured by the AIR system was demonstrated by looking at the data itself and noting that the AIR program would not record at night or during bad weather. 2 F.4th at 342. And in Martin, the Court analyzed the gaps in surveillance by examining how often the Flock system at issue captured an image of the defendant’s car, and how much time elapsed between those captures. 753 F. Supp. 3d at 472-73. Thus, even though CSLI data may help contextualize the gaps in the Flock surveillance data here, the need for context is not so great as to justify compelling Plaintiffs to turn over a detailed record of their movements derived from their cell phones. 2. To determine whether a surveillance system qualifies as a drag-net system, courts examine the data gathered by that system and what it reveals. To determine whether a surveillance system qualifies as a drag-net, Courts examine what the data provided by that system, combined with other information garnered through more traditional investigative techniques, shows the individuals who access it. See Leaders of a Beautiful Struggle, 2 F.4th at 342. Drag-net surveillance systems invade an individual’s reasonable expectation of privacy because the amount of data they collect about that person’s
movements provides the government with “an intimate window into [the] 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 (quoting Jones, 565 U.S. at 415 (Sotomayor, J., concurring)). Therefore, the courts faced with determining whether a particular system qualifies as a drag-net surveillance system consider whether the data, combined with other information, can provide the same “intimate window” into an individual’s life that CSLI data did in Carpenter. Id. In both Carpenter and Beautiful Struggle, the Supreme Court and the Fourth Circuit examined the surveillance systems’ ability to capture the “whole of individuals’ movements.” Leaders of a Beautiful Struggle, 2 F 4th at 333; Carpenter, 585 U.S. at 313. But in both cases, the holding ultimately rested on what the data provided by those systems could show, not whether they captured some specific percentage of a person’s movements. This reasoning is especially visible in Leaders of a Beautiful Struggle, where the Fourth Circuit emphasized that even with gaps in surveillance, the data was still enough to enable the government to make “deductions about ‘what a person does repeatedly, what he does not do, and what he does ensemble,” which “reveals more about a person than does any individual trip viewed in isolation.” 2 F.4th at 342 (quoting United States v. Maynard, 615 F.3d 544, 562-63 (D.C. Cir. 2010)). The fact that the data itself allowed the government to make these deductions is where the Fourth Amendment issue lies, and what takes the surveillance at issue beyond short-term surveillance that merely augments police capabilities. Jd. at 343. What the government can glean from the data it gathers, not whether that data is truly all-encompassing, is what the Court will eventually have to examine in this case. Courts analyzing Carpenter-based claims challenging the use of ALPR systems similarly look at what the data provided by those systems allowed the government to see.'? In Martin, the
'9 At the hearing, Defendants emphasized that the courts that have considered whether the use of ALPRs constitutes a search under the Fourth Amendment have done so in criminal cases, where there is generally 18
Court held that the Flock system at issue was not a drag-net surveillance system because “only three Flock cameras captured one picture each of the exterior of [the defendant’s car] .. . Out of the approximately 2,500 pictures that Officer Redford looked through on the Flock database, those are the only ones that captured [the defendant’s] vehicle in the 30-day timeframe in which Flock retains the data.” 753 F. Supp. 3d at 472. Examining this data, the Court noted that the three images could not demonstrate the route defendant allegedly took to go between the sites of two separate robberies, let alone the “totality of [his movements].” Jd. Ultimately, the Court held that “three individual snapshots of [the defendant’s] brief location at specific times hardly rise to the level of persistent, unceasing public surveillance that the courts found troublesome in Carpenter and Beautiful Struggle.” Id. at 473. Thus, the Court in Martin anchored its analysis in whether the data before it showed that the Flock system’s surveillance was so pervasive as to qualify as a drag-net system. See id. It did not attempt to compare the Defendant’s actual movements to the data; it merely looked at the data. Id. The Court’s discussion of the same issue in Sturdivant focused on the fact that “[e]ven in combination with information from other sources, the ALPR surveillance in this case did not generate enough data for agents to ‘catalogue every single movement of [the defendant’s] car.” Sturdivant, 2025 WL 1633754, at *10 (quoting Carpenter, 585 U.S. at 310). Again, the Court relied on the data, finding that “no amount of inference,” stemming from a combination of the ALPR-captured images and other information, “could create a ‘detailed log of [the defendant’s]
some evidence of the defendants whereabouts apart from the ALPR data. But even in the cases where the court may have had other location-corroborating evidence to examine—specifically, Martin, decided by this Court—the Court ultimately based its decision on its examination of the data, not a comparison of the data to the defendant’s actual movements.
movements’ from the data at issue here.” Jd. (quoting Carpenter, 585 U.S. at 310).2° The Court concluded its opinion in Sturdivant by warning that, although the record before it did not support a drag-net analysis under Carpenter, ALPR systems could grow to eventually capture enough data to qualify as a drag-net surveillance system. Jd. at *11.?' Even in this parting warning, the Court focused only on the extent of the data collected by the system, not the percentage of the defendant’s movements that data captured. Jd.””
20 As explained above, supra Section III.A., the Court briefly addressed the defendant’s comparison of his true whereabouts to what the ALPR data showed. This brief discussion is not dispositive as to the issue of importance for two reasons. First, Plaintiffs in this case indicated at the hearing that they do not intend to make similar arguments at trial and will instead rely on what experts argue the government could infer given the Flock system data in their possession and other information garnered through conventional surveillance techniques. Second, the Court in Sturdivant broadened its discussion beyond the specific instance the defendant described and again emphasized what the data showed: “Nor are the 26 data points at issue in this case dense enough to infer that Mr. Sturdivant stayed at a particular location overnight. Indeed, none was captured before 6:30 a.m. or after 5:30 p.m., and none associates his car with a particular residence.” 2025 WL 1633754, at *10. 21 Judge Bea’s concurrence in United States v. Yang likewise cautioned that ALPR surveillance could become so pervasive as to create Fourth Amendment concerns under Carpenter in the future. United States v. Yang, 958 F.3d 851, 864 (9th Cir. 2020) (Bea, J., Concurring). But again, he explained that when that threshold is crossed depends on the amount of data collected by the system and what that data can show, not whether the ALPR system at issue captures a certain percentage of an individual’s movements; “If the technology evolves in the way amici hypothesize, then perhaps in the future a warrant may be required for the government to access the [ALPR] database, but this should only be the case if the database evolves to provide comparable location information to the records at issue in Carpenter.” Id. 2 See also United States v. Cooper, No. CR 23-131, 2025 WL 35035, at *6 (E.D. La. Jan. 6, 2025) (“Ultimately, resolution of this issue focuses on ‘the extent to which a substantial picture of the defendant’s public movements are revealed by the surveillance’ from the ALPR.”) (quoting Commonwealth v. McCarthy, 142.N.E.3d 1090, 1104 (Mass. 2020)); United States v. Jiles, No. 8:23-CR-98, 2024 WL 891956, at *19 (D. Neb. Feb. 29, 2024) (“Because the data gathered by ALPRs is of such limited volume and utility, unlike GPS or cell-site location information, Deputy Parmer did not perform a ‘search’ within the meaning of the Fourth Amendment when he accessed the ALPR database.”); United States v. Porter, No. 21-CR- 00087, 2022 WL 124563, at *3 (N.D. Ill. Jan. 13, 2022) (“Under the framework outlined in United States v. Jones, the database query response did not reveal intimate details of Porter’s daily life, nor did it track his every movement . . . Rather, the query produced limited images of Porter’s vehicle at public locations over a period of approximately eight weeks, which reveals little about the intimate details of his life.”) (internal citations omitted) United States v. Rubin, 556 F. Supp. 3d 1123, 1129-30 (N.D. Cal. 2021) (“Although it is not clear precisely how many entries the NCRIC ALPR database contained for Rubin’s vehicle, there is no reason to believe that the database provided a detailed log of Rubin’s movements.”). United States v. Bowers, No. 2:18-CR-00292-DWA, 2021 WL 4775977, at *4 (W.D. Pa. Oct. 11, 2021) (“This limited data collection does not even begin to approach the same degree of information as that gathered in Carpenter, nor does it otherwise implicate similar privacy concerns.”). 20
Taken together, these cases demonstrate that the most important issue before the Court will be whether the data gathered by the Flock system, when combined with other information gathered in other ways, rises to the level of surveillance decried in Carpenter and Beautiful Struggle by providing an intimate window into the Plaintiffs’ life through their repeated movements. Indeed, at the hearing, Defendants agreed that the legal question before the Court is whether the data provided by the Flock system allows the government to learn the intimate details and associations of an individual’s life. Addressing this issue will involve examining the data gathered by the system, determining what inferences it allows, and then deciding if those inferences are so invasive as to infringe on a subjective and objective expectation of privacy. See Leaders of a Beautiful Struggle, 2 F.4th at 343. Finding out how many of the Plaintiffs’ trips the Flock system failed to capture is minimally relevant to answering that question in that it may demonstrate the technical capabilities of the Flock system and enable easier comparisons to other technologies. But it is not an important part of answering that question, given that precedent tasks the Court with examining the data itself and what it shows. The Court recognized, when denying Defendants’ Motion to Dismiss, that the remaining disputes of material fact included the extent to which Plaintiffs’ vehicles were photographed by the Flock system and “Defendants’ ability to effectively ‘track’ Plaintiffs’ vehicles as they move through Norfolk.” ECF No. 29 at20.n.10. These factual disputes, too, are resolved by, among other considerations, examining and analyzing the data at issue. 3. Compelling Plaintiffs to produce their CSLI data burdens their privacy interests. Compelling Plaintiffs to produce their CSLI data to Defendants, which offers a detailed record of their movements, see Carpenter, 585 U.S. at 300-301, raises privacy concerns. Defendants argued at the hearing that Plaintiffs mitigated their own privacy concerns by placing
their location data at issue when they filed suit, and that privacy concerns are sufficiently mitigated via the protective order entered in this case, ECF No. 34. But the Court is not persuaded that Plaintiffs must surrender their own privacy interests to vindicate their constitutional rights. Additionally, other courts have recognized that privacy concerns may factor into the proportionality analysis under Rule 26. See Williams v. United States, 331 F.R.D. 1, 4 (D.D.C. 2019) (“Rule 26(b)(1) provides that discovery should only be permitted to the extent that the likely benefit of such discovery outweighs its burden. Here, any minimal value defendants might derive from the location data does not outweigh the risks inherent in disclosure, and the Court’s privacy concerns remain largely unaddressed.”). Thus, these concerns provide more support to the conclusion that the request for Plaintiffs’ CSLI data is not proportional to the needs of the case. Defendants’ argument that the protective order entered in this case is sufficient to safeguard Plaintiffs’ privacy interests is not persuasive. Although the protective order prevents the disclosure of information to third parties, it does not stop the disclosure of that same information between the parties involved in litigation. That fact creates privacy concerns when, as is the case here, the government is the party seeking the information and urging the Court to rely on the safeguards of the protective order. Plaintiffs are concerned about the disclosure of their personal information to the government—here, the opposing party—so the protective order is irrelevant to that concern. In International Action Center v. United States, the Court addressed the same argument Defendants make here: As to the government’s first argument, that the parties are adequately protected with an order limiting the use and custody of the information provided in discovery to the parties, the government simply ignores the thrust of Plaintiffs’ fears. It is the government itself that Plaintiffs fear, and therefore confining that information to government . . . counsel will hardly assuage those fears or avoid the intimidation that Plaintiffs fear might follow.
207 F.R.D. 1, 3 (D.D.C. 2002). Precedent indicates that the percentage of Plaintiffs’ movements the Flock system captures is not sufficiently important to justify compelling Plaintiffs to produce their CSLI data. Plaintiffs’ privacy concerns also weigh against the limited utility and relevance of that data. Therefore, the Court finds that Defendants’ request for Plaintiffs’ CSLI data is not proportional to the needs of the case and DENIES the Motion to Compel.” Additionally, because the parties agree that Plaintiff Lee Schmidt’s cellphone service provider improperly disclosed CSLI records to Defendants after receiving a Rule 45 subpoena, ECF No. 97 at 3; ECF No. 98 at 2, the Court ORDERS Defendants to destroy the CSLI data they received in response to that subpoena. IV. CONCLUSION Defendants’ request for Plaintiffs’ CSLI data is only minimally relevant to the claims and defenses in the case and is not demonstrably important to the resolution of the issues before the Court. And granting the Motion to Compel would infringe on legitimate privacy concerns for Plaintiffs. Therefore, Defendants’ Motion to Compel, ECF No. 81, is DENIED, and the Court ORDERS Defendants to destroy the CSLI data they received from Plaintiff Schmidt’s cellphone service provider. The Clerk is DIRECTED to forward this Order to all counsel of record. It is so ORDERED. See! (X80 —Z0~ 7 Lawrence RK. Leonard United States Magistrate Judge Norfolk, Virginia August 15, 2025. 3 Because the Court denies Defendants’ Motion to Compel pursuant to Federal Rule of Civil Procedure 26, it does not need to reach the question of whether it may compel Plaintiffs to consent to the production of their CSLI data pursuant to 18 U.S.C. § 2702. 23