Date: Mar 19, 2026

Protective Sweeps of Vehicles After Driver’s Arrest Require Passenger-Specific Reasonable Suspicion of Dangerousness

Introduction

In United States v. Williams (10th Cir. Mar. 17, 2026), the Tenth Circuit addressed a recurring traffic-stop question: after officers arrest and handcuff a driver (eliminating the driver as an immediate threat), may officers nonetheless conduct a protective sweep of the vehicle based on a remaining passenger’s potential access to a weapon?

The stop involved Kylear Williams (driver) and Ajanaya Richardson (passenger and owner of the car). Officers learned Williams had warrants and a history involving violence, weapons, and gang ties; they arrested and removed him from the vehicle. They then removed and patted down Ms. Richardson—who was calm, cooperative, licensed, and not suspected of any crime—yet still swept the driver’s area and quickly found a loaded handgun and ammunition. Williams was indicted under 18 U.S.C. § 922(g)(1) (felon in possession of ammunition).

The key issue was whether the Fourth Amendment allowed the vehicle sweep on the theory that Ms. Richardson—based largely on her romantic relationship with Williams—was reasonably suspected to be armed and dangerous.

Summary of the Opinion

The Tenth Circuit reversed, holding the protective sweep was unreasonable under the Fourth Amendment. The court rejected the district court’s effectively per se rationale that a passenger becomes dangerous because she is the “romantic partner” of a gang-associated arrestee. Reasonable suspicion must be assessed under the totality of the circumstances, and the government did not identify passenger-specific, contemporaneous facts indicating Ms. Richardson was dangerous.

The court vacated Williams’s conviction and remanded.

Analysis

Precedents Cited

1) The protective-sweep framework: Terry extended to vehicles

Michigan v. Long, 463 U.S. 1032 (1983): The doctrinal anchor. Long permits a limited search of areas in a vehicle “in which a weapon may be placed or hidden” when officers reasonably suspect a person “poses a danger and may gain immediate access to a weapon.” The Williams majority treated Long’s two-part requirement—armed and dangerous—as independently meaningful and not satisfied by general association.
Terry v. Ohio, 392 U.S. 1 (1968): Long’s “specific and articulable facts” requirement is borrowed from Terry. Williams uses Terry to underscore that the justification must be fact-specific and not categorical.

2) Officer-safety principles in traffic stops

Rodriguez v. United States, 575 U.S. 348 (2015) (quoting Pennsylvania v. Mimms, 434 U.S. 106 (1983)): Recognizes the “legitimate and weighty” interest in officer safety during stops. Williams accepts that interest but insists it does not displace Long’s individualized suspicion requirements.
Maryland v. Wilson, 519 U.S. 408 (1997): Officers may order passengers out during a stop. Williams treats removal of the passenger as permissible, but distinguishes that authority from searching the vehicle.
Arizona v. Gant, 556 U.S. 332 (2009): Cited for the “countervailing interest” in constitutional protection. The majority also addresses a Gant concurrence to clarify that passenger-return access does not eliminate the need for “armed and dangerous” suspicion as to the person whose access creates the risk.

3) Tenth Circuit applications of Long and reasonable suspicion

United States v. Canada, 76 F.4th 1304 (10th Cir. 2023): Provides the opinion’s modern statement of the protective-sweep rule (“areas in which a weapon may be placed or hidden” when there is suspicion of danger and immediate access). Canada also supplies the standard caution: officer-safety sweeps are allowed, but they must fit the exception’s rationale. The Williams court viewed Canada as not resolving passenger-specific suspicion because Canada involved only a driver.
United States v. McGregor, 158 F.4th 1082 (10th Cir. 2025): Cited for the “totality of the circumstances” approach to reasonable suspicion for protective sweeps. Williams uses McGregor to reject a one-factor rule driven solely by relationship status.
United States v. Raban, 162 F.4th 1223 (10th Cir. 2025): Reiterates protective sweeps “exist for officer safety” and officers need not take unnecessary risks. Williams accepts that premise but holds the government must still show passenger-specific dangerousness.
United States v. Dennison, 410 F.3d 1203 (10th Cir. 2005): The government’s centerpiece. Dennison upheld a sweep where circumstances suggested a shared “enterprise” with a wanted passenger in a high-theft complex at 3 a.m. Williams distinguishes Dennison: there, officers developed suspicion that the driver himself was dangerous under the totality; here, the passenger-owner’s conduct and circumstances were non-threatening and lawful, and the district court treated relationship alone as enough.
United States v. Fager, 811 F.3d 381 (10th Cir. 2016): Mentioned by the district court but deemed not helpful; it involved a frisk of a driver’s person and did not address passenger-based vehicle sweeps.
Poolaw v. Marcantel, 565 F.3d 721 (10th Cir. 2009): A critical analogy. Poolaw held “a familial relationship is insufficiently particularized” to justify privacy intrusion. Williams extends that reasoning to romantic association, rejecting “criminality-by-association.”

4) Persuasive authority from another circuit

United States v. Johnson, 163 F.4th 933 (5th Cir. 2026): Highly influential in Williams’s reasoning. Johnson held that a protective sweep was unjustified where officers observed nothing suggesting the girlfriend-passenger was dangerous, even though she lived with a gang-member suspect and may have had a criminal record. Williams adopts Johnson’s insistence on contemporaneous indications of dangerousness, not mere relationship plus background.

5) Other citations framing standards and context

United States v. Canada, 76 F.4th 1304 (10th Cir. 2023); United States v. Pena, 115 F.4th 1254 (10th Cir. 2024); United States v. Johnson, 43 F.4th 1100 (10th Cir. 2022); United States v. Frazier, 30 F.4th 1165 (10th Cir. 2022): Provide the review framework (totality, deference to trained officers, de novo constitutional reasonableness, and the government’s burden).
Byrd v. United States, 584 U.S. 395 (2018): Cited for the proposition that vehicles are “effects” protected by the Fourth Amendment.
Wyoming v. Houghton, 526 U.S. 295 (1999): Mentioned below, but Williams finds it inapposite because it concerns evidence-based automobile search authority, not Long-style safety sweeps tied to a person’s dangerousness.

6) The dissent’s authorities

Pennsylvania v. Mimms, 434 U.S. 106 (1977): The dissent cites Mimms to emphasize traffic-stop dangers.
United States v. Samilton, 56 F.4th 820 (10th Cir. 2022); United States v. Fonseca, 744 F.3d 674 (10th Cir. 2014); United States v. Albert, 579 F.3d 1188 (10th Cir. 2009); United States v. Gurule, 935 F.3d 878 (10th Cir. 2019); United States v. Pettit, 785 F.3d 1374 (10th Cir. 2015): Cited for deference to officer inferences and common-sense evaluation.
Fishbein ex rel. Fishbein v. City of Glenwood Springs, 469 F.3d 957 (10th Cir. 2006): Used by the dissent to support an inference that a close associate who observes an arrest could become hostile and access weapons.

Legal Reasoning

Identifying the correct “person of concern.” Once Williams was arrested and handcuffed, the state’s remaining officer-safety justification for a Long sweep depended on the only person who might be allowed to reenter the car: Ms. Richardson. The majority therefore reframed the constitutional question: not whether officers reasonably suspected Williams was armed and dangerous (they did), but whether they reasonably suspected Ms. Richardson was armed and dangerous.
Separating “armed” from “dangerous.” The majority accepted that officers could infer Ms. Richardson might know about and have access to a weapon in her car (supporting the “armed/access” side). But it found the record failed on the “dangerous” side: the government offered no passenger-specific facts—conduct, statements, agitation, threats, evasiveness, intoxication, criminal suspicion, or other contemporaneous indicators—suggesting she posed a threat.
Rejecting “criminality-by-association.” The district court’s reasoning treated the romantic relationship as sufficient by itself—“the Court does not believe that the officers had to conclude through some type of observations outside her romantic relationship … that she was … a dangerous person.” The majority characterized this as an impermissible association test incompatible with totality-of-the-circumstances analysis and with Poolaw’s requirement of particularized suspicion.
Distinguishing Dennison’s “common enterprise” logic. Williams explains that Dennison involved much more suspicious circumstances (time, location, story inconsistencies, connection to an arrest-warrant passenger, and a scenario supporting inference that the driver himself was dangerous). By contrast, Ms. Richardson presented as a lawful owner, cooperative, calm, and not suspected of wrongdoing—facts the district court’s per se approach effectively discounted.
Using Johnson (5th Cir. 2026) as a benchmark for passenger-based sweeps. The majority found Johnson persuasive because it required some fact “contemporaneous to or arising out of” the arrest indicating the girlfriend-passenger was dangerous. Williams adopts that approach and notes Johnson was, if anything, a stronger case for the government than Williams (live-in girlfriend plus possible felony status), yet still failed.

Impact

1) No per se “romantic partner” dangerousness rule. Williams is best read as establishing that, in the Tenth Circuit, an occupant’s romantic relationship with an arrestee—standing alone—cannot supply reasonable suspicion that the occupant is “dangerous” for purposes of a Long protective sweep.

2) Passenger-based Long sweeps require passenger-specific facts. Officers may consider relationships, gang ties, and criminal history as part of the totality, but Williams requires additional, individualized, contemporaneous indicia that the passenger poses a danger (especially when the passenger is compliant, unaccused, and otherwise lawful).

3) Litigation focus shifts to “dangerousness” evidence. After Williams, suppression disputes in passenger cases will likely turn on articulation of specific facts demonstrating the passenger’s threatening conduct or other concrete reasons to fear imminent harm—not merely the arrestee’s background or generalized assumptions about loyalty.

4) Operational consequences for stops ending in driver arrest. When the driver is arrested and removed, police may still order the passenger out and take reasonable safety steps, but a vehicle sweep tied to the passenger’s future access must be justified with passenger-centered reasonable suspicion. Otherwise, evidence recovered may be suppressed.

Complex Concepts Simplified

Protective sweep (vehicle): A limited search of places in a car where a weapon could be hidden, justified not by probable cause of a crime but by officer safety under Michigan v. Long. It is narrower than a full evidentiary search.
Reasonable suspicion: A lower standard than probable cause. Officers must point to specific, articulable facts—plus reasonable inferences— showing the person is likely armed and dangerous. Hunches or stereotypes are not enough.
“Armed” vs. “dangerous”: Under Long, potential weapon access (“armed”) does not automatically prove intent or likelihood of violence (“dangerous”). Williams treats these as separate requirements.
Totality of the circumstances: Courts consider all facts together—time, location, behavior, statements, criminal indicators, and officer experience. Williams condemns collapsing that multi-factor inquiry into a single association-based shortcut.
Criminality-by-association: Treating a person as suspicious or dangerous merely because of their relationship to someone with a criminal history. Williams rejects this as inconsistent with individualized Fourth Amendment suspicion.

Conclusion

United States v. Williams tightens the fit between the officer-safety rationale and the scope of vehicle sweeps after a driver’s arrest: when the driver is secured, the government must justify any further Long sweep by showing passenger-specific reasonable suspicion that the remaining occupant is both armed (or can access a weapon) and dangerous. A romantic relationship—even to a gang-associated arrestee—cannot substitute for particularized facts under the totality of the circumstances. The decision thus reinforces the Fourth Amendment’s insistence on individualized suspicion while preserving Long’s core purpose: allowing protective action when concrete facts indicate real danger.