People v. Levy CA1/4
- Area
- criminal
- Panel
- Goldman (author), Janice R. Brown
Analysis
No analysis yet. When one exists it is a model's reading of this opinion — the rulings it reviewed, whether it agrees with the court, and the authorities it checked — shown beside the court's own disposition, never in place of it. How analyses are made.
Opinion
Filed 9/28/26 P. v. Levy CA1/4 NOT TO BE PUBLISHED IN OFFICIAL REPORTS California Rules of Court, rule 8.1115(a), prohibits courts and parties from citing or relying on opinions not certified for publication or ordered published, except as specified by rule 8.1115(b). This opinion has not been certified for publication or ordered published for purposes of rule 8.1115.
IN THE COURT OF APPEAL OF THE STATE OF CALIFORNIA
FIRST APPELLATE DISTRICT
DIVISION FOUR
THE PEOPLE, Plaintiff and Respondent, A172151 v. ALEXANDER LEVY, (County of San Mateo Super. Ct. No. 21SF006855A) Defendant and Appellant.
Federal law requires electronic service providers to report instances of apparent child sexual abuse material to the National Center for Missing and Exploited Children (National Center). (18 U.S.C. § 2258A(a), (b).) This case began when Dropbox, Inc., made such a report (known as a CyberTip), providing the means to access 13 videos that a Dropbox user had uploaded. The National Center conveyed the CyberTip to an intermediary, which in turn sent it to local law enforcement for the jurisdiction from which the Dropbox user had uploaded the videos. The officer who received the CyberTip viewed the videos without a warrant. Describing what he saw, he applied for search warrants that yielded evidence leading to Alexander Levy’s conviction for possessing child or youth pornography. Levy moved to suppress the evidence on the ground that the officer’s warrantless search violated the Fourth Amendment to the United States Constitution. Citing the private search doctrine, the magistrate, and then the trial court, denied Levy’s motions, finding that the videos were no longer private because a Dropbox worker had already viewed them. On appeal, Levy contends that the magistrate improperly relied on inadmissible hearsay in finding that a Dropbox worker had conducted a prior private search. Alternatively, he argues that the evidence is insufficient to support the magistrate’s finding that a Dropbox worker conducted such a search. He also argues that the officer’s viewing of the files exceeded the scope of any prior private search. We are unpersuaded by Levy’s arguments, and therefore affirm the judgment. BACKGROUND I. Sergeant Gordon Currie of the San Mateo County Sheriff’s Office investigated internet crimes against children. He had training about and experience receiving CyberTips. When a CyberTip included access to files that Currie “kn[e]w someone ha[d] looked at,” he would open and review the files himself. His purpose was to determine whether the image or video in a file constituted child sexual abuse material that could provide the basis of a criminal act. In 2020, Currie received the Dropbox-initiated CyberTip at issue. It included means to access 13 files identified as apparent child sexual abuse material. Each file contained a video that had been classified as A1, A2, B1, or B2 based on the perceived age range of the subject and the act shown. The categories A1 and A2 refer to prepubescent minors engaging in either a sex act (A1) or a lascivious exhibition (A2), while the categories B1 and B2 refer to pubescent minors engaging in a sex act (B1) or a lascivious exhibition (B2). The CyberTip classified five videos as A1, two as A2, five as B1, and one as 2 B2. For each file, the CyberTip stated “Yes” to the question “Did Reporting [electronic service provider] view entire contents of uploaded file?” Without a search warrant, Currie opened the files and viewed the videos, confirming that they contained child sexual abuse material. He then applied for search warrants directed at Google and Comcast because information in the CyberTip indicated that the person who had uploaded the videos to Dropbox had used those companies’ services. In his warrant affidavit, Currie described what he had seen in the videos. After the warrants issued, Google and Comcast provided information from which law enforcement identified Levy and his address. Based on information in the original warrant affidavit and on new information from warrant returns by Google and Comcast, Currie obtained three more sets of warrants, which permitted searches of Levy’s person, home, devices, and accounts with electronic service providers. The searches uncovered more child sexual abuse material, including hundreds of images on Levy’s phone and about 1,000 images and 2,000 videos in a cloud storage account. The district attorney charged Levy with one count of possession of child or youth pornography (Pen. Code, § 311.11, subd. (a))1 with special allegations of possessing over 600 images (§ 311.11, subd. (c)(1)) and of possessing sadomasochistic images (§ 311.11, subd. (c)(2)). II. Before his preliminary hearing, Levy moved to suppress all evidence obtained from the search warrants. (See § 1538.5, subd. (f).) He contended that Currie’s warrantless review of the 13 videos violated the Fourth 1 All further statutory references are to the Penal Code unless otherwise indicated. 3 Amendment and that the subsequent warrants all depended upon the unlawful initial review. The prosecution sought to establish that the private search doctrine allowed Currie to watch the videos without a warrant. Under that doctrine, a police search that merely replicates a prior private search does not violate the Fourth Amendment. (U.S. v. Jacobsen (1984) 466 U.S. 109, 115–120.) According to the prosecution, a private actor—i.e., a Dropbox worker—had reviewed the videos before submitting the Dropbox CyberTip report to the National Center. Currie testified about receiving the CyberTip and about his subsequent actions. Tobias Wulff, Dropbox’s content safety manager, also testified for the prosecution. He ran a team that reviewed and enforced Dropbox’s Acceptable Use Policy. Initially, the prosecutor presented him as a fact witness who would testify based on his personal knowledge of Dropbox’s policies regarding the review of files suspected of containing child sexual abuse material. But after he had described the policies and procedures purportedly in place at the time of the CyberTip at issue, Wulff testified that he joined Dropbox in 2022. He had no personal knowledge of Dropbox policies that were in place in 2020 or of the actions of the person who submitted the CyberTip. Levy objected to Wulff’s testimony on grounds of hearsay and foundation. The prosecution then shifted its position, offering Wulff as an expert in Dropbox’s policies rather than as a fact witness. Wulff testified that part of the training that Dropbox had provided him for his position as content safety manager had entailed reading documents that covered both historical and current policies. Those policies included the obligations of a Dropbox worker who reviewed a file that had been flagged for apparent child sexual abuse 4 material. Based on Wulff’s background, the magistrate accepted him as an expert in Dropbox’s policies regarding the reporting of such material. The relevant Dropbox policies and procedures, Wulff testified, had been the same in 2020 as they were in 2022. Dropbox workers screened the content of user files flagged for apparent child sexual abuse material. If they found such material, they would generate a CyberTip, designating what they saw as A1, A2, B1, or B2. Entry of such a designation on a CyberTip form would cause an autofill response of “Yes” to the form’s question “Did Reporting [electronic services provider] view entire contents of uploaded file?” Dropbox policy required a worker to view the entire content of a file—for example, an entire video—before entering a designation and submitting the form. When presented with the CyberTip at issue, Wulff recognized it as appearing to be a CyberTip. He testified that it was consistent with Dropbox procedures for workers who reviewed apparent child sexual abuse material. He identified parts of the report that Dropbox provided as part of its procedures, including the content designations for each file and information associated with the Dropbox user, such as an email address and IP addresses. Levy argued that the CyberTip document had not been properly authenticated and was hearsay. The prosecution did not offer any exception to the rule against hearsay that would permit the admission of the CyberTip for the truth of the statements it contained. The prosecution expressly declined to argue that the business record exception applied. Instead, the prosecution argued that it offered the CyberTip for the nonhearsay purpose of showing its existence—rather for the truth of statements it contained. Its existence, the prosecution argued, was evidence that a Dropbox employee had 5 viewed the 13 videos. Otherwise, under Dropbox policy, no CyberTip would have been generated and submitted. Levy also argued that People v. Sanchez (2016) 63 Cal.4th 665 (Sanchez) made inadmissible, as hearsay, Wulff’s expert testimony about Dropbox’s 2020 policies. That testimony, according to Levy, amounted to case-specific facts rather than general background information. The magistrate admitted the CyberTip into evidence “for the non- hearsay purposes of showing the effect of the document on Sgt. Currie to explain his subsequent action, and to inform and illuminate the probative value of expert testimony about the fact of an extant CyberTip.” The magistrate implicitly rejected Levy’s Sanchez argument. The magistrate found that, under Dropbox policies and procedures, a Dropbox worker must visually review the entire content of a file flagged for child sexual abuse material. If a reviewer does not find such material in a flagged file, Dropbox does not generate a CyberTip. The magistrate further found that the CyberTip at issue was, in fact, a CyberTip report; that it included 13 attached Dropbox files; and that it appeared to comply with Dropbox reporting policies and procedures. He concluded that the prosecution had demonstrated by a preponderance of the evidence that a Dropbox worker had conducted a private search of the 13 files before Currie conducted his warrantless search. And Currie’s search, the magistrate further concluded, did not exceed the scope of the Dropbox worker’s search. The magistrate therefore denied Levy’s motion to suppress and held him to answer.
6 III. In the trial court, Levy renewed his motion to suppress (§ 1538.5, subd. (i)) and moved to dismiss the count against him (§ 995). The parties did not submit new evidence. The trial court denied both motions. It found that the magistrate had properly admitted the CyberTip into evidence and that Wulff’s testimony about the CyberTip and Dropbox’s policies had not been offered for an inadmissible hearsay purpose under Sanchez, supra, 63 Cal.4th 665. That evidence, the court concluded, sufficed to establish a private search. Levy pled no contest to the possession count and admitted the special allegations and aggravating circumstances. The court sentenced him to two years in prison. DISCUSSION I. Before turning to Levy’s arguments, we address the standard of review and the private search doctrine. After a magistrate at a preliminary hearing denies a section 1538.5, subdivision (f)(1), motion to suppress, the defendant may renew the motion before the trial court under section 1538.5, subdivision (i). If, as here, the parties do not introduce new evidence, we review the magistrate’s suppression determination directly. (People v. Turner (2017) 13 Cal.App.5th 397, 404.) At an evidence suppression hearing, the prosecution bears the burden of proving admissibility by a preponderance of the evidence. (U.S. v. Matlock (1974) 415 U.S. 164, 177, fn. 14.) The magistrate “sits as the finder of fact with the power to judge credibility, resolve conflicts, weigh evidence, and draw inferences.” (People v. Shafrir (2010) 183 Cal.App.4th 1238, 1244.) We review the magistrate’s 7 rulings on the admissibility of evidence for abuse of discretion (People v. Daily (1996) 49 Cal.App.4th 543, 549–550); and we review the magistrate’s factual determinations for substantial evidence (People v. Turner, supra, 13 Cal.App.5th at p. 404). We independently determine whether, on such facts, a search violates the Fourth Amendment. (Ibid.) The Fourth Amendment protects people from unreasonable searches by the government. (U.S. v. Jacobsen, supra, 466 U.S. 109, 113–114.) Warrantless searches are presumptively unreasonable. (Katz v. U.S. (1967) 389 U.S. 347, 357.) But the Fourth Amendment does not apply to searches by private parties. (Jacobsen, at pp. 113–114.) A private party may conduct a search that would be unconstitutional if performed by the government. (See ibid.) Under the private search doctrine, after a private party conducts such a search, the government may replicate it without implicating the Fourth Amendment. (U.S. v. Jacobsen, supra, 466 U.S. at pp. 115–117.) “Once frustration of the original expectation of privacy occurs, the Fourth Amendment does not prohibit governmental use of the now nonprivate information.” (Id. at p. 117.) “The Fourth Amendment is implicated only if the authorities use information with respect to which the expectation of privacy has not already been frustrated.” (Ibid.) II. Levy contends that the magistrate twice erred by relying on hearsay evidence in determining that Dropbox conducted a prior search. Without the inadmissible evidence, Levy argues, no evidence supported the magistrate’s determination. We conclude that the magistrate did not abuse his discretion in overruling Levy’s hearsay objections. We address them in turn.
8 A. Levy attacks Wulff’s expert testimony about Dropbox’s historical policies and procedures as impermissibly conveying case-specific hearsay in violation of Sanchez, supra, 63 Cal.4th 665. On appeal, he does not challenge the magistrate’s acceptance of Wulff as an expert on Dropbox’s policies regarding the reporting of apparent child sexual abuse material. In Sanchez, the court recognized that an expert’s testimony regarding his or her general knowledge often rests on hearsay. (Sanchez, supra, 63 Cal.4th at pp. 675–676, 685.) The court “d[id] not call into question the propriety of an expert’s testimony concerning background information regarding his knowledge and expertise and premises generally accepted in his field.” (Id. at p. 685.) For example, in order to testify to accepted medical knowledge, a physician would not be required to “personally replicate all medical experiments dating back to the time of Galen” in order to testify to accepted medical knowledge. (Id. at p. 675; see also id. at p. 685.) Likewise, a gang expert could testify based on knowledge gathered from hearsay that a street gang had adopted a particular symbol and that a tattoo of that symbol showed gang membership. (Id. at p. 677.) But the court held that an expert’s testimony may not include case- specific hearsay. (Sanchez, supra, 63 Cal.4th at pp. 670, 684–686.) “Case- specific facts are those relating to the particular events and participants alleged to have been involved in the case being tried.” (Id. at p. 676.) “The expert is generally not permitted . . . to supply case-specific facts about which he has no personal knowledge.” (Ibid.) In Sanchez, for example, a gang expert could not testify about the defendant’s prior police contacts and activities of the defendant about which the expert had learned from other policer officers and their reports. (Id. at pp. 672–673, 694.)
9 The Supreme Court discussed the issue further in People v. Veamatahau (2020) 9 Cal.5th 16 (Veamatahau.) There, an expert in forensic testing of controlled substances opined that pills found on the defendant were alpraxoalm, otherwise known as Xanax. (Id. at pp. 22–23.) To identify the pills, he had used a database to search and compare the logos that appeared on them with reference images from the database. (Ibid.) That method of identification, he testified, was generally accepted in the scientific community. (Ibid.) He conceded that he did not know who put the markings on the pills found on the defendant. (Ibid.) That testimony did not consist of case-specific hearsay. (Veamatahau, supra, 9 Cal.5th at pp. 27–28.) The database contained hearsay, but not about the events in the defendant’s case. (Id. at p. 27.) Rather, it contained information about how any pills containing certain chemicals would appear. (Ibid.) The expert properly could apply that general background hearsay information to the case-specific facts of the markings on the defendant’s pills because the expert personally had observed them. (Id. at pp. 26–35.) Here, the Attorney General does not dispute that Wulff’s expert testimony about Dropbox’s 2020 policies and procedures rested on hearsay. Wulff did not work at Dropbox in 2020. He knew that the then-current policies and procedures had been in place for a long time because when he “onboarded and joined Dropbox, part of [his] training was [an] ample amount of reading those documents that cover both historical and current policies.” The documents contained out-of-court statements about the policies in effect in 2020 on which he relied for their truth. We agree with the Attorney General, however, that the hearsay was not case specific. Although Veamatahau is not directly analogous, it is instructive. Like the database in Veamatahau that concerned any pills 10 (Veamatahau, supra, 9 Cal.5th at p. 27), the documents relied on by Wulff concerned any Dropbox review of flagged child sexual abuse material—not the review that generated the CyberTip at issue in particular. Sanchez did not bar Wulff from testifying about the policies and procedures governing the workers who reviewed flagged files. (See Sanchez, supra, 63 Cal.4th at pp. 684–686.) We are not persuaded by Levy’s argument that Wulff’s testimony was case specific because it concerned “[a] document’s content on a particular date” and because it applied to a specific worker. Wulff’s testimony was broader than that. On this point, Veamatahau is again instructive. There, the hearsay database concerned all pills, including the defendant’s pills. (Veamatahau, supra, 9 Cal.5th at p. 27.) The inclusion of the defendant’s pills within the larger category of all pills did not render the expert’s reliance on the database case specific. (Id. at pp. 26–35.) Here, Wulff conveyed that the hearsay policies and procedures applied to all Dropbox reviewers for a long time. The inclusion of a specific reviewer on a specific date within that larger category likewise did not render Wulff’s reliance on the document case specific. The principle may be weaker here than in Veamatahau. There, scientists relied on and used the database for purposes other than the case against the defendant. Here, there is no such evidence about the use of Dropbox’s historical policies. But the reasoning in Veamatahau still tends to cast Wulff’s testimony as general rather than case specific. Citing People v. Valencia (2021) 11 Cal.5th 818 (Valencia), Levy also contends that expert testimony is case specific unless the subject matter is generally accepted as accurate by experts in the field. (See id. at p. 837 & fn. 15.) Wulff’s testimony, Levy argues, does not meet that standard.
11 Levy did not raise this argument in the proceedings before the magistrate. Consequently, the prosecution had no opportunity to cure the potential defect by introducing evidence that other experts in the field accepted the content of Wulff’s testimony as accurate. Even on appeal, Levy did not raise this issue until his reply brief, precluding the Attorney General from responding to it. In any event, we do not think that Valencia extends so far as Levy suggests. There, the court ruled that past predicate offenses—relevant in the context of gang sentencing enhancements—are case specific, and the prosecution must prove them with competent evidence. (Valencia, supra, 11 Cal.5th at p. 838.) A past predicate offense is part of a pattern of criminal activity by a gang to which a defendant belongs. (Id. at p. 829.) An expert with no personal knowledge may not testify about the particulars of such an offense based solely on conversations with other officers or on police reports. (Id. at pp. 827, 838–839.) The Attorney General argued that Sanchez applied only to “ ‘the particular events and participants alleged to have been involved in the case being tried’ ”—and therefore not to past offenses by other perpetrators. (Id. at pp. 838–839, quoting Sanchez, supra, 63 Cal.4th at p. 676.) But the court disagreed. In reaching its conclusion, the court discussed the distinction between case-specific facts and background information. (Valencia, supra, 11 Cal.5th at pp. 835–838.) The court noted that “experts are given latitude over lay witnesses [to testify based on hearsay] only to the extent they are conveying acquired expertise in their field.” (Id. at p. 835.) “[I]f experts give testimony that goes beyond their own experience or beyond principles generally accepted in their field, the justifications for allowing greater evidentiary latitude cease to apply.” (Id. at p. 836.) The court described Sanchez and 12 Veamatahau as using the terms “ ‘general background information’ and ‘case- specific facts’ to distinguish . . . between hearsay that may be admitted because it is generally accepted by experts in the field, and facts that cannot be proven by hearsay because that reliability justification is absent.” (Id. at p. 837; see also id. at p. 837, fn. 15.) We do not take this language in Valencia to mean that general information becomes case specific simply because there is no evidence that the information is generally accepted by experts in the field. Imagine that a physician testifies, contrary to a consensus in the medical field, that arsenic is known to prolong life. The problem is not that the physician’s hearsay testimony is case specific. The problem is that it is not reliable, as shown by the lack of support from the majority of other experts in the field. Here, Levy does not argue that Wulff’s testimony is unreliable because there is no evidence that its content is generally accepted by other experts. We acknowledge the mismatch between Wulff’s status as an expert and the type of testimony the prosecution elicited from him about the policies in place in 2020. As Levy points out, the prosecution initially offered Wulff as a fact witness. He started by testifying about 2020 policies based on his personal knowledge of 2022 policies. Only after Levy objected on foundation and hearsay grounds did the prosecution seek to have him accepted as an expert. Likely as a result, Wulff does not appear to be a typical expert working within a typical field of expertise. As highlighted by Valencia, there is no evidence about how many other experts are in Wulff’s field, what shared background establishes them as experts, or whether they agree that Dropbox’s policies did not change from 2020 to 2022. Nonetheless, we do not agree with Levy that Wulff smuggled case-specific facts into evidence under the guise of general background.
13 B. Levy also contends that the magistrate should not have considered the CyberTip because it is hearsay and because Wulff could not testify about its content without violating Sanchez, supra, 63 Cal.4th 665. The magistrate erred, Levy contends, by admitting the CyberTip into evidence for limited purposes, including “to inform and illuminate the probative value of expert testimony about the fact of an extant Cybertip.” Hearsay evidence is an out-of-court statement “offered to prove the truth of the matter stated.” (Evid. Code, § 1200, subd. (a).) It is inadmissible for that purpose unless an exception to the rule applies. (Id., § 1200, subd. (b).) But “[i]f a fact in controversy is whether certain words were spoken or written and not whether the words were true, evidence that these words were spoken or written is admissible as nonhearsay evidence.” (People v. Fields (1998) 61 Cal.App.4th 1063, 1068.) Here, the CyberTip’s out-of-court statements included that Dropbox had viewed the entire contents of each of the 13 listed videos files and that each video contained a specific type of child sexual abuse material, designated as A1, A2, B1, or B2. The prosecution did not offer the CyberTip under any exception to the hearsay rule. Therefore the magistrate could not consider the document for the truth of those statements. Nor could Wulff testify, based on those statements, that a Dropbox worker had reviewed the content of the 13 files. This would be analogous to the gang expert in Sanchez impermissibly testifying about the defendant’s police contacts based solely on hearsay in police reports. (See Sanchez, supra, 63 Cal.4th at pp. 672–673, 694.) But that is not what happened. Wulff testified generally about CyberTips and the Dropbox procedures that generate them. When asked if 14 he recognized what the CyberTip document at issue was, he testified, “Yes. That’s exactly—it looks like a CyberTip report.” The prosecutor asked him whether the CyberTip complied with Dropbox’s policies and procedures. She elaborated, in response to an objection that the CyberTip document was produced by the National Center, that she was asking whether the information included in the report matched the information that Dropbox policies required workers to include. Wulff answered, “Yes.” He did not testify that a Dropbox worker had opened the 13 files or reviewed the videos that they contained. His testimony did not depend on a Dropbox worker having opened or reviewed the videos in this particular case. Sanchez therefore did not prevent the magistrate from considering the testimony. We are not persuaded by Levy’s argument that “the only way that [the CyberTip] could comply with Dropbox policies and procedures is if the content of the [document] is true.” As the magistrate recognized, the accuracy or truth of the information in the document is not the issue. The magistrate did not consider the document as direct evidence that a Dropbox worker had viewed the 13 videos. Rather, it was circumstantial evidence of such a search, strengthened by its consistency with Dropbox policy. The evidentiary value was that the CyberTip contained the appropriate type of information required by Dropbox—not whether the information was accurate. III. Levy argues that even if Wulff’s testimony about Dropbox’s policies and procedures is admissible, the magistrate erred in the way that he used it. Specifically, Levy contends that the testimony does not support an inference that a reviewer at Dropbox actually reviewed the 13 videos. We disagree. Wulff’s testimony supplied substantial evidence supporting the magistrate’s findings that Dropbox policy requires a worker to review the 15 contents of a file flagged for child sexual abuse material and that Dropbox does not generate a CyberTip unless, during that review, the worker finds apparent child sexual abuse material. Based on these findings, the magistrate reasonably could infer from the existence of the CyberTip that a Dropbox worker had opened and reviewed the 13 files. Levy cites Juen v. Alain Pinel Realtors, Inc. (2019) 32 Cal.App.5th 972 (Juen) for the proposition that the existence of a policy is not evidence that a worker followed the policy. Juen is distinguishable. There, the defendants moved to compel arbitration in a putative class action involving real estate transactions. (Juen, supra, 32 Cal.App.5th at p. 975.) They submitted a copy of a listing agreement signed by both the plaintiff and a defendant real estate agency. (Id. at p. 976.) The agreement’s arbitration clause required both parties’ initials, but the submitted copy of the agreement contained only the plaintiff’s. (Id. at pp. 976, 979.) The agency had, pursuant to a document-retention policy, destroyed the original listing agreement. (Id. at p. 976.) The defendants sought to establish that the real estate agency had initialed the original listing agreement. (Juen, supra, 32 Cal.App.5th at p. 976.) The agency’s managing broker submitted a declaration describing an agency policy in place at the time the parties entered the agreement. (Ibid.) The policy required agents to present an executed listing agreement to the managing broker if the client had initialed the arbitration clause. (Ibid.) The managing broker would then initial the arbitration clause on behalf of the agency. (Ibid.) The managing broker’s declaration did not state that the managing broker had a habit of accounting for every listing contract executed by the agent responsible for the plaintiff’s listing agreement. (Id. at p. 981.) The agent did not submit a declaration stating that she had presented the 16 managing broker with all executed agreements she procured, or that she had presented the managing broker with the specific executed agreement at issue. (Ibid.) The court of appeal upheld the trial court’s conclusion that the defendants had failed to prove that the managing broker had initialed the arbitration clause. (Juen, supra, 32 Cal.App.5th at p. 981.) In discussing its decision, the court stated that “the existence of a policy is not evidence of adherence to the policy, and the . . . defendants offered no evidence (directly or circumstantially through custom or habit) showing that [the agent] had actually presented [the] plaintiff’s listing agreement to [the managing broker].” (Id. at p. 980.) Levy leans, in particular, on the first part of this statement. Juen and the present case both involve the significance of a workplace policy. In each case, the policy instructs that (1) an input (2) triggers a required worker process (3) that results in a particular output. In Juen, the input consisted of the executed listing agreement with an arbitration clause initialed by the client; the process consisted of the agent presenting the agreement to the managing broker and the managing broker initialing the arbitration clause; and the resulting output consisted of an executed listing agreement with the arbitration clause initialed by both parties. (See Juen, supra, 32 Cal.App.5th at p. 976.) In this case, the input consisted of a file with apparent child sexual abuse material; the process included a Dropbox worker viewing the file’s entire content; and the resulting output consisted of a CyberTip report that contained information about the file, including a content designation. In both cases, direct evidence supported the existence of the workplace policy and the input.
17 In this case, however, there is also direct evidence of the existence of the usual output of the process—while in Juen there was not. Here, the question is whether Dropbox’s policies are circumstantial evidence that, in conjunction with the CyberTip’s existence, permit an inference that a Dropbox worker performed the required process to generate a CyberTip. In Juen, the question was whether the existence of the policy alone sufficed to permit an inference that the missing output—a mutually initialed form—had been created. Here, unlike in Juen, the output—the CyberTip—is not missing. This additional piece of circumstantial evidence in this case fills in the final step of the process and, with reference to the policy, strengthens the inference that the middle step occurred. In light of this key difference, we disagree with Levy that Juen’s statement applies. In addition, Juen is procedurally distinguishable. There, the court of appeal noted that it did not review the trial court’s conclusion for substantial evidence because the trial court had held that the defendants, who bore the burden of proof on the disputed issue, had not carried that burden. (Id. at pp. 978–979.) When “ ‘the issue on appeal turns on a failure of proof at trial, the question for a reviewing court becomes whether the evidence compels a finding in favor of the appellant as a matter of law. [Citations.] Specifically, the question becomes whether the appellant’s evidence was (1) “uncontradicted and unimpeached” and (2) “of such a character and weight as to leave no room for a judicial determination that it was insufficient to support a finding.” ’ ” (Ibid.) Under this standard, the court of appeal held only that “the evidence presented in the [managing broker’s declaration] is not of such character and weight as to compel a finding that [she] initialed the listing agreement as a matter of law.” (Id. at p. 981.) Here, by contrast, the magistrate found that the People had carried their 18 burden of proof. We therefore review the magistrate’s factual findings only to determine whether the People submitted substantial evidence sufficient to support those findings, not whether they submitted evidence that compelled such findings as a matter of law. Our standard of review is thus more deferential than that which the procedural posture of Juen required. We conclude that the magistrate did not err by relying on Wulff’s testimony about Dropbox’s policies and procedures as circumstantial evidence that a Dropbox worker reviewed the 13 videos. We also conclude that substantial evidence, including the testimony, supports the magistrate’s finding that a Dropbox worker viewed the 13 videos. IV. Levy argues in the alternative that the magistrate erred by denying the motion to suppress because Currie’s viewing of the 13 videos exceeded the scope of the Dropbox worker’s search. We disagree. Under the private search doctrine, a government search is constitutional if it does not exceed the scope of a prior private search. (U.S. v. Jacobsen, supra, 466 U.S. 109, 115.) The question is whether the government’s search uncovered anything different from or in addition to what the earlier private search had revealed. (See id. at p. 120.) Citing U.S. v. Wilson (9th Cir. 2021) 13 F.4th 961 (Wilson), which also involved a CyberTip, Levy contends that Currie’s search exceeded that of the Dropbox worker because Dropbox communicated only part of the information that Currie would learn when he viewed the videos. Levy reasons that watching a video is entirely different from reading Dropbox’s designations. “Currie’s search of the 13 files,” Levy argues, “allowed the government to glean specific details unapparent from the basic classifications . . . .”
19 In Wilson, the Ninth Circuit held that the private search doctrine did not justify the government’s warrantless search of images attached to the defendant’s email. (Wilson, supra, 13 F.4th at p. 964.) Google’s automated system had flagged the attached images and had generated a CyberTip that the National Center forwarded to local law enforcement. (Ibid.) An officer opened and viewed the images without a warrant. (Ibid.) The officer relied on descriptions of what he viewed to obtain warrants to search the defendant’s email account and his home. (Ibid.) No human at Google had opened the attached email files and viewed the images. (Ibid.) Google’s system had determined that the images were the same as ones Google employees had already viewed and classified as child pornography. (Ibid.) One ground for the court’s decision was “the large gap between the information in the CyberTip and the information the government obtained and used to support the warrant application . . . .” (Wilson, supra, 13 F.4th at p. 973.) Viewing the attachments, the court explained, “substantively expanded the information available to law enforcement far beyond what the label alone conveyed, and was used to provide probable cause to search further and to prosecute.” (Ibid.) The officer’s search had exceeded Google’s search, the court concluded, because it allowed the government “to learn new, critical information.” (Id. at pp. 971–972.) This case is readily distinguishable from Wilson because there, no human opened the defendant’s email attachments or viewed the iterations of the images they contained. (Wilson, supra, 13 F.4th at p. 964.) The court expressly stated that the officer’s search exceeded the scope of Google’s search because “the government saw more from its search than the private party had seen.” (Id. at p. 974.) It also stated that “[a]lthough Google’s proprietary technology labelled Wilson’s email attachments as ‘A1,’ ” Google 20 did not view the content of the images, “as no Google employee had opened and viewed the attachments . . . .” (Ibid.) “The critical fact,” the court explained, “is that no Google employee viewed [the defendant’s] files before [the officer] did.” (Ibid.) Here, substantial evidence supports the magistrate’s finding that a Dropbox worker viewed the 13 videos before Currie did. Wilson’s reasoning therefore does not apply. DISPOSITION The judgment is affirmed. GOLDMAN, J.
WE CONCUR:
BROWN, Acting P. J. GORDON, J.* * Judge of the Superior Court of California, City and County of San Francisco, assigned by the Chief Justice pursuant to article VI, section 6 of the California Constitution. 21
Cites, in this corpus
- United States v. Luke Wilson 18-50440 · 2021-09-21
- People v. Valencia S250218 · 2021-07-01
- People v. Veamatahau S249872 · 2020-02-27
- People v. Sanchez S216681 · 2016-06-30
- People v. Shafrir A125880 · 2010-03-29
- People v. Fields B112055 · 1998-02-27
- People v. Daily B094022 · 1996-09-18