United States v. Tew: Cloud-Account Warrants Must Be Meaningfully Particularized, Yet Officers May Receive Leon Good-Faith Protection; Severance Motions Are Due When the Basis Is “Reasonably Available,” Not “Definitively Available”

Court: United States Court of Appeals for the Tenth Circuit
Date: April 6, 2026
Disposition: Judgments affirmed (convictions and sentences for Michael Aaron Tew and Kimberley Ann Tew)

1. Introduction

United States v. Tew consolidates two appeals arising from a multi-year fraud against National Air Cargo (“National”), a government contractor. Michael Tew, National’s CFO (formerly a contractor through his company Sand Hill), and his wife Kimberley Tew participated in a scheme—assisted by National’s comptroller, Jonathan Yioulos—to submit falsified invoices for nonexistent vendors and route payments to themselves and intermediaries, ultimately draining over $5 million. At trial, Yioulos testified as a cooperating witness, and the spouses pursued blame-shifting defenses that became openly antagonistic in openings and closings.

The consolidated appeals presented two sets of issues:

  • Fourth Amendment / Stored Communications Act warrant: Kimberley challenged a warrant served on Apple, Inc. seeking extensive cloud-account content (emails, messages, photos, location/access data), arguing lack of particularity and facial overbreadth.
  • Severance: Both defendants argued the district court erred in denying severance in light of antagonistic defenses—Kimberley via pretrial and “supplemental” trial assertions, Michael via a mid-trial motion after Kimberley’s opening statement.

The Tenth Circuit issued a notable, technology-forward Fourth Amendment holding: the Apple cloud warrant was insufficiently particularized, but suppression was avoided because the Government carried its burden under the United States v. Leon good-faith exception. Separately, the court tightened Rule 12 practice around severance: a severance motion must be filed when the basis is “reasonably available,” and the district court’s timeliness determination is reviewed for abuse of discretion.

2. Summary of the Opinion

2.1 Fourth Amendment

  • The court held the Apple cloud-account warrant violated the Fourth Amendment’s particularity requirement because it authorized seizure of broad categories of data limited mainly by very broad “Target Offenses” (including 18 U.S.C. § 371 and wire fraud statutes), with “including” language that did not meaningfully confine seizure to the National fraud scheme.
  • Nevertheless, the court held the evidence was admissible under the good-faith exception because the warrant was not so “plainly” deficient that no reasonable officer could rely on it, especially given a magistrate’s approval, the magistrate’s signature on the warrant application/affidavit, prosecutorial review, investigative context, and the lack of on-point cloud-warrant guidance.

2.2 Severance

  • Kimberley: Her on-appeal severance theory was treated as unavailable due to failure to seek a ruling on the trial-stage antagonism theory and failure to argue plain error/good cause; even on the merits, she did not show prejudice sufficient to outweigh jury instructions and the record (including an acquittal on one count showing separateness of consideration).
  • Michael: His mid-trial severance motion was untimely under Rule 12 because the basis was “reasonably available” before trial. The court adopted abuse-of-discretion review for timeliness determinations (a question it characterized as effectively first impression in the circuit). In any event, the court found the defenses were not “mutually exclusive” under Tenth Circuit severance law; they amounted to “finger pointing.”

3. Analysis

3.1 Precedents Cited

A. Particularity in digital searches and the anti–general-warrant principle

  • United States v. Palms and United States v. Otero anchor the Tenth Circuit’s modern particularity framework: warrants must describe items to be seized with specificity that leaves “nothing to the discretion” of executing officers (Palms), and the expansion of digital storage makes particularity “that much more important” (Otero).
    Tew extends this logic from local devices (computers/phones) to cloud-based accounts, emphasizing that cloud accounts can intermingle and expose even more personal data than a standalone device.
  • Maryland v. Garrison is invoked (via Otero) for the Founding-era purpose of particularity: preventing “exploratory searches.” The opinion repeatedly frames the Apple warrant as reminiscent of general warrants in effect, if not in form.
  • Coolidge v. New Hampshire supplies the vocabulary of “exploratory rummaging,” which Tew uses to characterize the privacy dangers of broad cloud-account disclosure followed by government filtering.
  • United States v. Carey is used for the proposition that a computer search is categorically different from a physical file cabinet search—an analogy Tew then builds upon to say that cloud-account searches are categorically different again from device searches.
  • United States v. Suggs and United States v. Leary supply the operative formulation: warrants must be as specific “as the government’s knowledge and circumstances allow” (Suggs quoting Leary), and “an unadorned reference to a broad federal statute does not sufficiently limit” a warrant (Leary).
    In Tew, the “Target Offenses” included the broadest kinds of criminal prohibitions (notably 18 U.S.C. § 371 and wire fraud statutes), and that breadth was central to the court’s conclusion that the warrant’s “limitations” were illusory.
  • United States v. Cooper and United States v. Suggs govern when an affidavit can cure a warrant’s lack of particularity. Tew applies the strict two-part requirement: (1) attachment and (2) express incorporation. Because neither was satisfied for the warrant itself (only for the application), the court refused to use the affidavit to narrow the warrant.
  • United States v. Wagner is addressed and distinguished. The Government relied on Wagner to argue that a warrant can be sufficiently limited “as a whole” even if each category does not repeat the same limitation. Tew rejects this move because only one exemplar paragraph tied the seizure to the National fraud scheme, so the warrant “as a whole” did not replicate Wagner’s contextual narrowing.
  • United States v. Porter provides the interpretive tool for the warrant’s repeated use of “including.” Tew treats “including” as non-exhaustive, meaning the paragraph referencing “National Air Cargo” did not actually confine the broader seizure authorization.

B. Supreme Court privacy cases shaping the digital context

  • Riley v. California frames the qualitative difference between cell phones and physical containers and underscores how digital content can expose a “revealing montage” of a life. Tew uses Riley to emphasize that an Apple cloud account can be “cell phone and then some” (emails, messages, photos, location data, device history across multiple devices).
  • Carpenter v. United States is deployed for its insistence that the Fourth Amendment must respond to new technology enabling “detailed, encyclopedic, and effortlessly compiled” surveillance; while Carpenter concerned warrant necessity, Tew uses its logic to heighten concern about broad retrospective access to location/access data.
  • Kyllo v. United States appears (through Carpenter) for the principle of preserving the “degree of privacy” that existed at the Founding despite technological change.

C. Cloud/social-media warrant analogies from other courts

  • United States v. Scully is cited to describe the now-familiar two-step Stored Communications Act process: provider disclosure followed by government review/seizure.
  • United States v. Zelaya-Veliz is used both to confirm Rule 41(e)(2)(B) recognition of the two-step process and to highlight privacy problems when warrants lack time limits in social media contexts.
  • United States v. Blake supplies an Eleventh Circuit warning about warrants compelling disclosure of “virtually every type of data” in a social media account. Tew treats these cases as persuasive but stresses Apple accounts can contain even more intimate, cross-platform data than typical social media accounts.
  • United States v. Nejad is cited for the practical point that cyber/financial-crime evidence can be difficult to specify ex ante, but Tew holds the need for flexibility does not justify a warrant as broad as this one.

D. Wire-fraud breadth and “broad statute” limits

  • Kousisis v. United States (quoting Pasquantino v. United States) is used to underscore that wire fraud’s language is “undeniably broad,” reinforcing why mere statutory references do not meaningfully cabin a warrant.
  • Roche v. United States appears (via Leary) as an example of a warrant found overbroad when limited only by mail fraud.

E. Good faith and when it fails

  • United States v. Leon supplies the good-faith exception’s core.
  • Mapp v. Ohio frames the exclusionary rule’s deterrent purpose.
  • United States v. Cardall establishes the presumption of good faith when a neutral magistrate issues the warrant.
  • United States v. Knox (quoting Leon) emphasizes that executing officers are not expected to second-guess a magistrate’s technical sufficiency determinations.
  • United States v. Leary provides the objective test: whether a reasonably well-trained officer would know the search was illegal despite the magistrate’s authorization.
  • United States v. Riccardi supplies the contextual inquiry into whether officers could “reasonably presume” validity, and it informs the court’s weighing of execution circumstances.
  • United States v. Russian identifies circumstances that support good faith (e.g., magistrate also signing application/affidavit; same officer drafting and executing; investigative context).
  • United States v. Otero reappears for the relevance of consulting counsel in drafting warrants, a factor favoring good faith.
  • United States v. Corral-Corral is cited for the principle that some warrants are so deficient that reliance is “wholly unwarranted.”
  • United States v. Santiago (quoting Groh v. Ramirez) sets the high bar for denying good faith when a warrant “plainly” lacks particularity; Tew distinguishes Santiago (no crime named; catchall “any other information”) and Groh (no items described at all) as more glaring than this warrant.

F. Severance: standards, preservation, and timeliness

  • United States v. Pursley supplies the Tenth Circuit’s three-step severance inquiry (mutual exclusivity; threat to specific trial right or reliable verdict; judicial economy balancing).
  • Zafiro v. United States is cited (through Pursley) for the rule that mutually antagonistic defenses are not prejudicial per se.
  • United States v. Jones (quoting United States v. Dazey) states the demanding mutual-exclusivity standard: believing one core defense must require disbelieving the other.
  • United States v. Linn is invoked for the characterization of mere “finger pointing” as insufficient.
  • United States v. Hargrove supports the presumption juries follow limiting instructions; Tew bolsters this with the verdict’s differential treatment of the defendants on one money-laundering count.
  • U.S. Aviation Underwriters, Inc. v. Pilatus Bus. Aircraft, Ltd. and United States v. Leffler are used to frame preservation: an issue must be presented and a ruling sought; otherwise it is forfeited, and failure to argue plain error can convert forfeiture into waiver for appellate purposes.
  • United States v. Herrera and United States v. Bowline add an extra constraint: the Tenth Circuit will not review untimely Rule 12 severance arguments absent “good cause.”
  • On timeliness and standard of review:
    • The court treated the “reasonably available” timing question under Fed. R. Crim. P. 12(b)(3)(D) as effectively first impression in the circuit.
    • It adopted abuse-of-discretion review, citing United States v. Burrage by analogy and relying on the district court’s superior position to evaluate evolving case dynamics.
    • It imported the abuse-of-discretion definition from F.D.I.C. v. Oldenburg and the reversal threshold from Nalder v. W. Park Hosp.

3.2 Legal Reasoning

A. Why the Apple warrant failed particularity

The court’s particularity analysis turned on four interlocking conclusions:

  • No affidavit cure: Because the affidavit was not attached to and expressly incorporated into the warrant itself, United States v. Suggs foreclosed using it to narrow the warrant. This choice matters in digital cases: affidavits often contain the very “guardrails” (known accounts, co-conspirator identifiers, targeted date windows) that prevent a warrant from functioning as a general warrant.
  • Structure and “including” breadth: Reading the warrant “practically and commonsense[ly]” (Suggs), the court treated the “including” clauses as illustrative rather than limiting (United States v. Porter). The only paragraph explicitly tied to “a conspiracy or scheme to defraud National Air Cargo” did not confine the rest of the seizure authorization, and references to “Target Offenses” were too broad to do meaningful limiting work.
  • Overbroad reliance on broad statutes: The warrant’s principal limiter was a list of “Target Offenses” that included very broad crimes (18 U.S.C. § 371; wire fraud; conspiracy to commit fraud). Under United States v. Leary, “unadorned reference” to broad statutes is insufficient; Tew emphasizes this is especially true when the “place” searched is a cloud account containing comprehensive personal data.
  • Ambiguous date limitation and privacy spillover: Even the “since May 1, 2018” language was ambiguous (created-after vs. evidencing-after), and the warrant’s permission to seize “motive” evidence risked sweeping in intimate marital communications untethered to the National fraud. The court found this especially troubling given the Government had information in the affidavit that could have narrowed scope but did not.

The resulting doctrinal move is important: Tew does not hold cloud-account warrants are invalid per se; rather, it demands that the warrant’s text itself contain meaningful narrowing principles commensurate with the privacy stakes of cloud aggregation.

B. Why suppression was denied anyway: Leon good faith in an under-developed doctrinal area

Despite finding a constitutional defect, the court applied the good-faith exception because the warrant was not “plainly” unconstitutional in the Groh v. Ramirez/United States v. Santiago sense. The opinion stressed that this was “a close call,” but it credited several circumstances:

  • Judicial approval plus contextual cues: The magistrate judge signed the warrant, the application, and the affidavit, making it more reasonable for officers to assume the warrant aligned with probable cause limits described in supporting materials (a factor recognized in United States v. Russian).
  • Prosecutorial review: The application indicated an Assistant U.S. Attorney reviewed it—relevant not as neutrality, but as evidence the agents sought legal input (consistent with United States v. Otero and United States v. Riccardi factors).
  • Investigative continuity: The executing agent did not draft the affidavit, but she drafted other warrant materials in the same investigation—suggesting awareness of the investigation’s scope and constraints.
  • Protective protocol: The record showed a taint protocol to protect privileged materials, an additional indicator of constitutional-minded execution.
  • Doctrinal uncertainty around cloud warrants: The court emphasized the lack of “specific guidance” in authoritative precedent about properly limiting cloud-account warrants of this sort. That absence made it harder to say a “reasonably well trained officer” would know reliance was unlawful under Leary/Leon.

A notable methodological point appears in the footnote: the court deliberately chose to decide both particularity and good faith, rather than skipping to good faith, because the law around cloud-account particularity is thin and the issue was fully briefed. This signals a precedential intent: to guide future warrant drafting even while affirming convictions.

C. Severance: preservation, timeliness, and the Pursley merits framework

Kimberley

Kimberley’s appellate severance theory failed primarily on procedural grounds:

  • Her pretrial severance motion was framed around Bruton v. United States and Confrontation Clause concerns, but her on-appeal theory focused on antagonistic defenses under United States v. Pursley.
  • At trial, counsel said she was “supplement[ing]” the severance record but did not actually request a ruling on a Pursley severance motion, triggering forfeiture principles under U.S. Aviation Underwriters, Inc. v. Pilatus Bus. Aircraft, Ltd..
  • She did not argue plain error in her opening brief, leading to waiver under United States v. Leffler, and she did not demonstrate “good cause,” which United States v. Herrera and United States v. Bowline require before the Tenth Circuit will review untimely severance issues.

The court also indicated that, even on the merits, the jury’s differential verdict (acquitting Kimberley on one money laundering count while convicting Michael) supported the presumption from United States v. Hargrove that jurors follow instructions to consider evidence separately.

Michael

Michael squarely moved for severance after Kimberley’s opening statement, preserving the issue substantively but failing procedurally under Rule 12:

  • Reasonably available, not definitively available: The district court reasoned—and the circuit approved—that Rule 12(b)(3)(D) does not require certainty; it requires filing when the basis is reasonably available. Given years of mutual inculpation (proffers), prior conflict warnings, Kimberley’s earlier severance posture, and discovery showing money flowed through Michael’s accounts, the antagonism was reasonably foreseeable.
  • Standard of review: The Tenth Circuit adopted abuse-of-discretion review for the Rule 12 timeliness determination and found no abuse.
  • No good cause: Michael’s argument that a “non-frivolous claim of prejudice” equals good cause was rejected as logically nullifying Rule 12’s timeliness requirement.

The court nonetheless explained why the motion would fail on the merits under Pursley:

  • The defenses were not “mutually exclusive” under United States v. Jones/United States v. Dazey; they amounted to blame-shifting/finger-pointing under United States v. Linn.
  • Mutually antagonistic defenses are not prejudicial per se under Zafiro v. United States and Pursley.
  • Judicial economy strongly favored a single trial given the already-seated jury and years-long pretrial history; severance mid-trial would require mistrial, new trial settings, and re-litigation of pretrial matters.

3.3 Impact

A. Drafting and litigating cloud-account warrants in the Tenth Circuit

Tew is a substantial signal to investigators and magistrate judges: cloud-account warrants—especially those seeking broad content plus metadata like location/access history—must be drafted with tight, text-based limiting principles that appear on the face of the warrant (or via a properly attached and incorporated affidavit).

Practical consequences likely include:

  • Increased use of incorporated attachments that specify the suspected scheme/victim(s), known co-conspirators, and narrower data types.
  • Greater emphasis on date bounds that are unambiguous (e.g., “data created between X and Y” and/or “data evidencing specified offenses committed between X and Y”).
  • More frequent inclusion of search protocols or filtering constraints for highly private repositories, especially where warrants reach photos, location data, and multi-device histories.

Yet the court also provided the Government a litigation off-ramp: in novel cloud-warrant contexts, Leon good faith may still salvage evidence, at least until precedent (including Tew itself) makes deficiencies “plain.” In other words, Tew simultaneously raises the constitutional drafting bar and narrows immediate suppression consequences where agents can show reasonable reliance in a developing area.

B. Severance practice: earlier filing and clearer preservation

On joinder/severance, the case underscores two practice points:

  • Rule 12 timing: Anticipated antagonism must be raised pretrial if reasonably foreseeable; waiting for “proof” in openings may be too late.
  • Preservation: “Making a record” is not enough—counsel must move for relief and seek a ruling on the specific legal theory later pressed on appeal.

4. Complex Concepts Simplified

  • Particularity (Fourth Amendment): A warrant must be specific about what the government is allowed to take. The point is to prevent a “general warrant” that lets agents search broadly for anything that looks suspicious.
  • Cloud-account warrant “two-step” process: The provider (Apple) first discloses a large set of data; then the government reviews it to decide what it is authorized to “seize” and use as evidence. The privacy risk is that disclosure itself can be massive, so the warrant’s limits must be real.
  • Good-faith exception (Leon): Even if a warrant is later found unconstitutional, evidence may still be used if officers reasonably relied on the judge-issued warrant. The question is not whether the warrant was bad, but whether it was so obviously bad that no reasonable officer would rely on it.
  • Severance: Trying co-defendants together is common. Severance is granted only when a joint trial threatens a concrete trial right or makes the verdict unreliable. Defendants pointing fingers at each other usually is not enough.
  • “Reasonably available” under Rule 12: Motions that must be filed before trial (including severance) are due when a competent lawyer could foresee the basis—not when the harm becomes undeniable in front of the jury.
  • Forfeiture vs. waiver (appeal posture): If counsel does not properly raise an issue and obtain a ruling, the issue is typically lost on appeal unless counsel argues “plain error” and meets a demanding standard.

5. Conclusion

United States v. Tew advances Fourth Amendment doctrine into the cloud era by holding that an Apple cloud-account warrant—authorizing review of vast personal data—failed the particularity requirement when limited largely by broad “Target Offenses” and non-limiting “including” language. At the same time, it illustrates the continuing strength of United States v. Leon in technologically evolving domains: where deficiencies are not “plain” and officers can show reasonable reliance on a magistrate-approved, prosecutor-reviewed warrant in a sparse-precedent area, suppression may be denied.

On criminal procedure, Tew also sends a clear message about severance: Rule 12 requires pretrial litigation when the basis is reasonably foreseeable, and appellate courts will not rescue untimely or improperly preserved severance claims—particularly when the claimed antagonism is merely “finger pointing” rather than genuinely mutually exclusive defenses under United States v. Pursley, United States v. Jones, and Zafiro v. United States.