¶ 1 Courtney Weakland appeals from her convictions for aggravated driving under the influence (DUI) while impaired to the slightest degree and aggravated driving with a blood alcohol concentration of .08 or more. She contends the trial court should have suppressed the results of blood-alcohol testing because police officers obtained her blood sample without a warrant and without valid consent. Although we agree the sample was unlawfully obtained, we conclude the officers acted in good faith, and exclusion of the evidence therefore was not required.

Factual and Procedural Background

¶ 2 "In reviewing the denial of a defendant's motion to suppress, we consider only 'evidence presented at the suppression hearing and view the facts in the light most favorable to sustaining the trial court's ruling.' " State v. Valenzuela , 239 Ariz. 299, ¶ 3, 371 P.3d 627 (2016) ( Valenzuela II ), quoting State v. Hausner , 230 Ariz. 60, ¶ 23, 280 P.3d 604 (2012). In February 2015, an Oro Valley police officer arrested Weakland for DUI, handcuffed her, put her in the back seat of his patrol car, and read her an "admin per se" form, pursuant to A.R.S. § 28-1321, which provided that Arizona law "require[d]" her to complete certain tests to determine her blood-alcohol concentration. Weakland submitted to blood testing.

¶ 3 Weakland was indicted on one count of aggravated DUI while impaired to the slightest degree and one count of aggravated DUI with a blood-alcohol concentration of .08 or more. Before trial, she moved to suppress all of the evidence acquired through the warrantless search and seizure of her blood, arguing the requirement language in the admin per se admonition coerced her consent. The court summarily denied her motion.

¶ 4 Following conviction on both counts, the trial court sentenced Weakland to concurrent four-month prison terms followed by concurrent five-year probationary terms. She timely appealed, and we have jurisdiction pursuant to A.R.S. §§ 13-4031 and 13-4033.

Discussion

¶ 5 On appeal, Weakland argues the blood test results should have been suppressed "because the police obtained her blood sample without a warrant and without valid consent." And she maintains the good-faith exception to the exclusionary rule, recognized by our supreme court in Valenzuela II , 239 Ariz. at 371, ¶¶ 31-35, 371 P.3d 627, does not apply. "We review the denial of a motion to suppress evidence for abuse of discretion, considering the facts in the light most favorable to sustaining the ruling." Id. ¶ 9. An error of law may constitute such an *448abuse. Id. We review de novo the applicability of the good-faith exception. State v. Havatone , 241 Ariz. 506, ¶ 11, 389 P.3d 1251 (2017).

¶ 6 The state implicitly concedes that Weakland's consent for the warrantless blood draw was involuntary, and therefore invalid, pursuant to Valenzuela II , 239 Ariz. 299, ¶ 33, 371 P.3d 627. Indeed, the record shows the officer's interaction with Weakland did not comply with the standards set forth in that decision, which directs officers to inform suspects of the provisions of the admin per se statute "in a way that does not coerce consent by stating or implying that officers have lawful authority, without a warrant, to compel samples of blood, breath, or other bodily substances." Id. ¶ 28. The state has not asserted any other basis to find the search permissible under the Fourth Amendment. Thus, the sole issue on appeal concerns whether the good-faith exception to the exclusionary rule applies.

¶ 7 "The exclusionary rule, which allows suppression of evidence obtained in violation of the Fourth Amendment, is a prudential doctrine invoked to deter future violations." Id. ¶ 31. Because it functions solely for that purpose, it is unwarranted if it "fails to yield 'appreciable deterrence.' " Davis v. United States , 564 U.S. 229, 236-37, 131 S.Ct. 2419, 180 L.Ed.2d 285 (2011), quoting United States v. Janis , 428 U.S. 433, 454, 96 S.Ct. 3021, 49 L.Ed.2d 1046 (1976). Furthermore, while "deterrent value is a 'necessary condition for exclusion,' ... it is not 'a sufficient' one." Id. at 237, 131 S.Ct. 2419, quoting Hudson v. Michigan , 547 U.S. 586, 596, 126 S.Ct. 2159, 165 L.Ed.2d 56 (2006). "[T]he deterrence benefits of suppression must [also] outweigh its heavy costs." Id.

¶ 8 "Therefore, when law enforcement officers 'act with an objectively reasonable good-faith belief that their conduct is lawful,' deterrence is unnecessary and the exclusionary rule does not apply." Valenzuela II , 239 Ariz. 299, ¶ 31, 371 P.3d 627, quoting Davis , 564 U.S. at 238, 131 S.Ct. 2419. In such cases, the good-faith exception to the exclusionary rule applies. Davis , 564 U.S. at 238-40, 131 S.Ct. 2419. But the exception will not apply when "officers exhibit deliberate, reckless, or grossly negligent disregard for Fourth Amendment rights," nor when circumstances show "recurring or systemic negligence." Havatone , 241 Ariz. 506, ¶ 21, 389 P.3d 1251, quoting Herring v. United States , 555 U.S. 135, 144, 129 S.Ct. 695, 172 L.Ed.2d 496 (2009). And "[t]he State bears the burden of proving the good-faith exception applies." Id. ¶ 19.

¶ 9 Weakland argues first that the state failed to raise the good-faith exception in the trial court, and has therefore waived any argument that it should apply. But "[w]e are required to affirm a trial court's ruling if legally correct for any reason and, in doing so, we may address the state's arguments to uphold the court's ruling even if those arguments otherwise could be deemed waived by the state's failure to argue them below."2 State v. Boteo-Flores , 230 Ariz. 551, ¶ 7, 288 P.3d 111 (App. 2012).

¶ 10 Weakland next contends the good-faith exception should not apply because "the state continued to disregard Arizona case law that established the admin per se warning misstates the law." Citing Carrillo v. Houser , 224 Ariz. 463, 232 P.3d 1245 (2010), she contends our courts had "established that the admin per se warning misstated the law" and that allowing the state to rely on the exception in view of that uncertainty in the law would "incentivize rather than deter unlawful police behavior."

¶ 11 In Valenzuela II , however, our supreme court determined that the good-faith exception applied in circumstances nearly *449identical to those before us here. The court determined that, as in this case, a DUI suspect's consent to a warrantless blood draw had been coerced by the officer's reading of the admin per se admonition. Valenzuela II , 239 Ariz. 299, ¶ 33, 371 P.3d 627. The court expressly rejected the argument Weakland makes as to Carrillo , explaining that case had held an arrestee must "unequivocally manifest assent to the testing by words or conduct" before an officer could proceed. Valenzuela II , 239 Ariz. 299, ¶ 34, 371 P.3d 627, quoting Carrillo , 224 Ariz. 463, ¶ 19, 232 P.3d 1245. The court had not, it stated, "suggested that the admonition ... was coercive." Id. Furthermore, the court expanded, it had not "ever questioned or overruled" cases concluding that Arizona's consent law required suspects to submit to blood testing, as stated in the admonition. Id. And, it pointed out that "our courts have continued to approve the admonition." Id .,citing State v. Valenzuela , 237 Ariz. 307, ¶ 24, 350 P.3d 811 (App. 2015) ( Valenzuela I ), vacated , 239 Ariz. 299, ¶ 36, 371 P.3d 627, and State v. Oliver , No. 2 CA-CR 2014-0359, ¶¶23-25, 2015 WL 4924747 (Ariz. App. Aug. 18, 2015) (mem. decision). Our supreme court having rejected this argument, we are bound to do so as well. See State v. Stanley , 217 Ariz. 253, ¶ 28, 172 P.3d 848 (App. 2007).

¶ 12 In her reply brief, however, Weakland also asserts that our supreme court's decision in State v. Butler , 232 Ariz. 84, 302 P.3d 609 (2013), which was issued between the arrest at issue in Valenzuela II and Weakland's arrest, set forth "unequivocally ... that notwithstanding the implied consent law, the validity of the consent must be determined solely upon the Fourth Amendment" and therefore "clarified" that the admonition was coercive. We disagree.

¶ 13 In Butler , our supreme court addressed "whether the Fourth Amendment to the United States Constitution requires that a juvenile arrestee's consent be voluntary to allow a warrantless blood draw." Id. ¶ 1. In that case, sixteen-year-old Tyler drove his car to school after smoking marijuana. Id. ¶¶ 2-3. The investigating officer read him the implied consent admonition, and he agreed to a blood draw, which he later challenged on the ground that his consent had been involuntary and that he lacked legal capacity to consent. Id. ¶¶ 4-5. The court acknowledged that, pursuant to the Supreme Court decision in Missouri v. McNeely , 569 U.S. 141, 133 S.Ct. 1552, 185 L.Ed.2d 696 (2013), a compelled blood draw is a search. Butler , 232 Ariz. 84, ¶ 10, 302 P.3d 609. But it noted that voluntary consent can allow a warrantless search. Id. ¶ 13. The court further noted that consent pursuant to § 28-1321"must be express," Butler , 232 Ariz. 84, ¶ 16, 302 P.3d 609, and that the statute "does not always authorize warrantless testing of arrestees," as some may "refuse by declining to expressly agree to take the test," in which case the statute requires a warrant, id. ¶ 17, quoting Carrillo , 224 Ariz. 463, ¶ 10, 232 P.3d 1245.

¶ 14 The court then concluded Tyler's consent had not been voluntary based on the totality of the circumstances, including his age, his demeanor, his having been placed in handcuffs, the length of his detention, and the absence of his parents. Id. ¶¶ 20-21. The court also noted that a deputy had "read the implied consent admonition to Tyler, once verbatim and once in what the deputy termed 'plain English,' concluding with the statement, 'You are, therefore, required to submit to the specified tests.' " Id. ¶ 20. The court did not address whether the language of the admonition was coercive, did not separately or clearly address its effect, and did not expressly raise a question as to its validity. Rather, it discussed the voluntariness of Tyler's consent under the totality of the circumstances-which, as set forth above, included numerous coercive factors-and, on that basis, it concluded Tyler's consent had been involuntary. Id. ¶¶ 18, 20-21.

¶ 15 Indeed, in Valenzuela II , the court limited its reliance on Butler to the propositions that a blood draw is a Fourth Amendment search and that it had in the past "focused on the totality of the circumstances, including but not limited to an officer's reading of an admin per se form, in determining whether a DUI suspect's consent to search was freely and voluntarily given." Valenzuela II , 239 Ariz. 299, ¶¶ 10, 20, 371 P.3d 627. It then took the additional step to apply law relating to claims of lawful authority, see *450Bumper v. North Carolina , 391 U.S. 543, 88 S.Ct. 1788, 20 L.Ed.2d 797 (1968), and to conclude that officers must inform suspects of the implied consent law "in a way that does not coerce consent," Valenzuela II , 239 Ariz. 299, ¶¶ 22-23, 28, 371 P.3d 627.

¶ 16 Thus, in determining whether the good-faith exception applied, the court in Valenzuela II stated that Arizona courts had not "questioned or overruled" existing law relating to the admin per se admonition, but had instead "continued to approve" it. Id. ¶ 34. We therefore conclude we must follow our supreme court's guidance and apply the good-faith exception in this case. As our supreme court determined in Valenzuela II , binding precedent supported the conclusion that consent given pursuant to the admin per se statute was voluntary and had not been overturned. 239 Ariz. 299, ¶ 34, 371 P.3d 627 ; cf. State v. Mitchell , 234 Ariz. 410, ¶ 31, 323 P.3d 69 (App. 2014) ("Thus, we do not suggest that law enforcement is expected to anticipate new developments in the law.").

¶ 17 Accordingly, the circumstances here differ from those in Havatone , in which our supreme court concluded the state "should have known that routinely directing blood draws from DUI suspects who were sent out of state for emergency treatment, without making a case-specific determination whether a warrant could be timely secured, was either impermissible or at least constitutionally suspect." 241 Ariz. 506, ¶ 20, 389 P.3d 1251. The court pointed out in that case, which addressed blood draws from unconscious suspects pursuant to § 28-1321(C), that "warrantless blood draws from DUI suspects based on a 'per se exigency' rather than the totality of individual circumstances have been discredited for over fifty years." Havatone , 241 Ariz. 506, ¶¶ 12, 24, 389 P.3d 1251, citing Schmerber v. California , 384 U.S. 757, 770, 86 S.Ct. 1826, 16 L.Ed.2d 908 (1966). Thus, the court determined that as to such blood draws, no binding precedent had ever approved the practice; the law relating to that portion of the statute had remained unsettled, in contrast to the law relating to the admin per se admonition that the court addressed in Valenzuela II . Havatone , 241 Ariz. 506, ¶ 28, 389 P.3d 1251.

¶ 18 Although courts have declined to extend the good-faith exception to cases "in which the appellate precedent, rather than being binding, is (at best) unclear," the exception applies when " 'binding appellate precedent' expressly instructed the officer what to do." United States v. Lara , 815 F.3d 605, 613 (9th Cir. 2016). Our supreme court concluded in Valenzuela II that there was express authority for an officer to rely on the admin per se admonition to obtain valid consent. Indeed, the court cautioned against "fault [ing] law enforcement for failing to anticipate that [the court] would disapprove the admin per se form" based on case authority that was "not dispositive of the issue," noting Arizona cases that had "continued to approve the admonition" before Valenzuela II was decided. 239 Ariz. 299, ¶ 34, 371 P.3d 627. Application of the exclusionary rule is thus inappropriate. See United States v. Johnson , 457 U.S. 537, 565, 102 S.Ct. 2579, 73 L.Ed.2d 202 (1982) ("[A] deterrence purpose can only be served when the evidence to be suppressed is derived from a search which the law enforcement officers knew or should have known was unconstitutional under the Fourth Amendment."). This is particularly so because the deterrent effect of exclusion, which is small when law enforcement could not have reasonably foreseen a change in the law, must outweigh its cost. See Davis , 564 U.S. at 236-37, 131 S.Ct. 2419.

¶ 19 The law does not require law enforcement to make legal assessments our courts have not made. At the time of Weakland's arrest, the courts of this state had not concluded the admin per se admonition was coercive, ineffective, or otherwise negated consent after Butler . See Valenzuela I , 237 Ariz. 307, ¶¶ 17-18, 350 P.3d 811, vacated , 239 Ariz. 299, ¶ 36, 371 P.3d 627 ; Oliver , No. 2 CA-CR 2014-0359, ¶¶ 10-13, 23-25 (relying on Valenzuela I and concluding admonition did not render consent involuntary); cf. State v. Okken , 238 Ariz. 566, ¶¶ 12, 17-24, 364 P.3d 485 (App. 2015) (citing Butler , but continuing to rely on Carrillo ); State v. Pena , No. 2 CA-CR 2013-0241, 2014 WL 3407343 (Ariz. App. July 10, 2014) (mem. decision) (citing Butler for various principles but not for bright-line rule that admonition impermissibly coercive);

*451State v. Figueroa , No. 2 CA-CR 2012-0458, 2014 WL 287959 (Ariz. App. Jan. 24, 2014) (mem. decision) (citing Butler and concluding consent to blood draw not coerced when officer read admin per se admonition).3 We therefore cannot say law enforcement was negligent in continuing to use the admonition.

¶ 20 Our dissenting colleague argues we cite only decisions that follow Valenzuela I "under principles of stare decisis." And he contends " Butler could be correctly cited for the proposition that the implied consent admonition was not sufficient alone to facilitate voluntary consent." He further maintains we have not "articulate[d] comprehensively the weight of that jurisprudence," asserting that only the decisions in Valenzuela I and II "squarely addressed whether the language of the implied consent admonition was compatible with constitutionally valid consent." Based on those two decisions, he argues only a minority of jurists accepted the state's position that, after Butler but before Valenzuela II , the implied consent admonition was sufficient to secure consent.

¶ 21 Even were we to accept our dissenting colleague's calculations as to the number of jurists who accepted the ongoing validity of the implied consent admonition after Butler , we would still be in a position of requiring law enforcement agencies to foresee a position that at least some jurists did not themselves adopt. Furthermore, we are aware of no authority to support the implicit suggestions that a case decided by a less than unanimous court carries less authority or that the number of jurists concurring in controlling decisions is relevant to good faith analysis.

¶ 22 In any event, his analysis of the post- Butler cases is incorrect. The court in Okken did, as our colleague describes, address whether the implied consent statutory scheme rendered consent involuntary and did not, as did Valenzuela II , directly address the admonition. But the court discussed Butler , stated that the statute required "actual consent before a warrantless search may be performed," discussed whether the statutory scheme was coercive, and yet made no mention of problems with the admonition-noting only that in Butler , "other factors," including "the defendant's age, his mental state, and the duration and circumstances of his detention," had rendered his consent involuntary. Okken , 238 Ariz. 566, ¶¶ 18-19 & n.1, 364 P.3d 485. Thus, the court apparently read Butler not as concluding that consent premised on the admonition was insufficient, but as determining that consent pursuant to the admonition could be rendered involuntary by other factors.

¶ 23 Furthermore, in Pena , our dissenting colleague and another member of this panel rejected a claim that a "warrantless breath test should have been suppressed under ... the Fourth Amendment."4 No. 2 CA-CR 2013-0241, ¶¶ 5, 10. Citing Butler , the decision identified various factors to determine voluntary consent, noting, as evidence supporting a finding of voluntariness, that Pena had been arrested before and that the officers had not displayed their weapons. Id. ¶¶ 7-8. The court then considered that officers had advised Pena of his rights pursuant to Miranda5 and given him the implied consent admonition. Id. ¶ 8. After initially refusing, Pena agreed to the test. Id. ¶¶ 4, 8. The decision does not address the validity of the admonition, but rather cites it as a basis for valid consent. Id. ¶ 8. Likewise, in Figueroa , a member of this panel rejected a claim of involuntary consent. No. 2 CA-CR 2012-0458, ¶¶ 18-21. In that case, we cited Butler as requiring compliance with the Fourth Amendment as to blood draws, rejected a *452claim that force had been used against the defendant, and stated that an officer had read the admonition to the defendant, who had then consented. Id. ¶¶ 19-20. In each of these cases, had this court understood Butler in the manner in which our dissenting colleague proposes law enforcement should have, we could not have reached a conclusion that the defendants' consent, obtained in response to the admonition, was voluntary.

¶ 24 Thus, it was not until our supreme court set forth in Valenzuela II the proper procedures for giving the admin per se admonition that law enforcement had a clear directive that officers could not continue to use the admonition to imply they had authority to compel a warrantless blood draw. We will not hold law enforcement to a standard that requires them to have foreseen that change. See Mitchell , 234 Ariz. 410, ¶ 31, 323 P.3d 69. The trial court therefore did not abuse its discretion in denying the motion to suppress. See Boteo-Flores , 230 Ariz. 551, ¶ 7, 288 P.3d 111 (appellate court will affirm if trial court legally correct for any reason).

Disposition

¶ 25 Weakland's convictions and sentences are affirmed.

¶ 26 In Valenzuela II , our supreme court addressed whether law enforcement agencies could have believed in good faith that a suspect's acquiescence to the "implied consent" admonition was sufficient to secure consent-valid under Fourth Amendment standards-for a warrantless blood draw. 239 Ariz. 299, ¶¶ 31-35, 371 P.3d 627. That analysis necessarily focused on whether binding precedent at the time of Valenzuela's blood draw would have placed those agencies on notice that "consent given solely in acquiescence to the admonition" might not be constitutionally valid. Id. ¶ 33. There, the court identified the then-pertinent precedents as Campbell v. Superior Court In and For Maricopa, 106 Ariz. 542, 546, 479 P.2d 685 (1971), and State v. Brito , 183 Ariz. at 535, 539, 905 P.2d 544 (1995) ; cases that it concluded "sanctioned use of the admonition." Valenzuela II , 239 Ariz. 299, ¶¶ 32-33, 371 P.3d 627. Accordingly, the Court held that the state acted in good faith. Id. ¶ 35.

¶ 27 Here, we address the same question with reference to markedly different binding precedent: Ms. Weakland was arrested in 2015, several years after the time of Mr. Valenzuela's arrest. Id. ¶ 4 (Valenzuela arrested in August 2012). Between those arrests, our supreme court issued its opinion in Butler , 232 Ariz. 84, 302 P.3d 609. That case rejected, for the first time, the state's theory that it could "imply" constitutionally valid consent from the suspect's decision to drive on Arizona's roads. Id. ¶¶ 17 -18 ; see Valenzuela II , 239 Ariz. 299, ¶ 25, 371 P.3d 627 (citing Butler to reject state's argument that consent could be implied from driving in Arizona). And, in applying that new understanding, the court's reasoning plainly suggested that the implied consent admonition was not alone adequate to secure voluntary consent. Butler , 232 Ariz. 84, ¶ 20, 302 P.3d 609.

¶ 28 Because Butler thus announced a marked departure from our state's prior precedent addressing the procedure for a constitutionally valid warrantless blood draw, I cannot agree with my colleagues that Valenzuela II resolves the application of the good-faith exception here. Instead, we address a novel question in this case that Valenzuela II did not entertain: whether, after Butler , law enforcement agencies could have continued to believe in good faith that acquiescence to the implied consent admonition was sufficient to demonstrate voluntary consent under Fourth Amendment standards.

¶ 29 In so doing, we must be mindful of the correct legal standards in applying the good-faith exception. The state bears the burden of establishing that it violated the defendant's constitutional rights in good faith. Havatone , 241 Ariz. 506, ¶ 19, 389 P.3d 1251. And, as our supreme court has clarified, it does not meet that burden by merely showing that the unconstitutional act had not yet been expressly forbidden. Id. ¶ 29. Rather, when binding precedent predating the police practice renders that practice one of "dubious constitutionality" or when the pertinent precedent addressing the practice is manifestly "unsettled," the logic of the exclusionary *453rule applies and the evidence secured by unconstitutional means must be suppressed. Id. (explaining that application of the good-faith exception in such circumstances would give "law enforcement officials ... little incentive to err on the side of constitutional behavior"). Thus, if case law, existing at the time of the draw of Weakland's blood, cast doubt on whether acquiescence to the advisory equated to constitutionally valid consent, the state is not entitled to the benefit of the good-faith exception.

¶ 30 We must therefore confront what Butler held. There, the state argued "that every Arizona motorist gives 'implied consent' under § 28-1321 and the tests administered under the statute are [therefore] not subject to a Fourth Amendment voluntariness analysis." Butler , 232 Ariz. 84, ¶ 9, 302 P.3d 609. Specifically, the state maintained that Arizona's implied consent law, as described in the advisory, either "constitutes an exception to the warrant requirement or satisfies the Fourth Amendment's requirement that consent be voluntary." Id. ¶ 17.

¶ 31 Our highest court bluntly characterized these arguments as "unconvincing" and clarified how it would view the application of the Fourth Amendment to DUI blood draws in future cases: "We hold now that, independent of § 28-1321, the Fourth Amendment requires an arrestee's consent to be voluntary to justify a warrantless blood draw." Id. ¶ 18 (emphasis added).

¶ 32 As the court's own language in that holding suggests, this upended prior understandings of blood draw procedure in Arizona DUI cases. Previous cases had expressly or implicitly accepted that the implied consent process itself satisfied all Fourth Amendment concerns. See Valenzuela II , 239 Ariz. 299, ¶¶ 32-33, 371 P.3d 627 (analyzing pre- Butler cases). Indeed, by holding otherwise in Butler , the supreme court reversed our opinion, which had stated, in conformity with prior Arizona jurisprudence, that "the informed consent statute presents no Fourth Amendment issue." State v. Butler, 231 Ariz. 42, n.6, 290 P.3d 435 (App. 2012) (citing previous Arizona cases accepting this principle).

¶ 33 In so reversing, the supreme court expressly held that an officer's compliance with Arizona's implied consent law was not sufficient alone to establish constitutionally valid consent to a blood draw: "Contrary to the State's argument, a compelled blood draw, even when administered pursuant to § 28-1321, is a search subject to the Fourth Amendment's constraints." Butler , 232 Ariz. 84, ¶ 10, 302 P.3d 609, citing McNeely, 569 U.S. at 144, 133 S.Ct. 1552 (compelled blood draw taken pursuant to Missouri's implied consent law subject to Fourth Amendment's restrictions on warrantless searches). Further, the court emphasized that a defendant's consent to any warrantless blood draw would be assessed under traditional constitutional standards for voluntariness. Id. ¶ 13.

¶ 34 As the state's arguments in Butler demonstrate, Arizona law enforcement agencies had persistently relied on the DUI suspect's acquiescence to the § 28-1321 advisory (the "implied consent admonition")6 as the exclusive basis for conducting a warrantless blood draw. To such agencies, the holding of Butler should have raised a very prominent red flag. Until Butler , the implied consent admonition had never been tasked with independently demonstrating a suspect's consent to a warrantless search under traditional Fourth Amendment voluntariness standards. Rather, as its content plainly demonstrates, the admonition had been designed to advise suspects of their legal obligation to submit to blood testing under Arizona's implied consent law-and to secure their acquiescence to that requirement.

¶ 35 After the supreme court's holding in Butler , any agency seeking to comply with evolving constitutional standards should have logically evaluated whether an advisory designed for one purpose (securing compliance with Arizona implied consent law) would coincidentally fulfill another (demonstrating voluntary consent to a search). For the reasons that follow, that inquiry would not have been reassuring.

*454¶ 36 First, in Butler , the supreme court's reasoning plainly demonstrated that submission to the admonition would not alone suffice to demonstrate voluntary consent. 232 Ariz. 84, ¶¶ 20-21, 302 P.3d 609. Indeed, the court affirmed the trial court's conclusion that the suspect, Tyler, had not voluntarily consented to the blood draw. Id. ¶ 21. It did so notwithstanding its recognition that Tyler had submitted to the admonition. Id. ¶ 20. That admonition, it pointedly observed, had been both read and explained to him. Id . In short, Butler not only squarely held that state implied consent law did not constructively establish constitutionally valid consent; it unsurprisingly found that the tool designed for advising a defendant of that law, the implied consent admonition, was inadequate alone to secure voluntary consent. Id.

¶ 37 Arguably, the court's reasoning went further than that. In a paragraph devoted to itemizing the reasons that "sufficient evidence supports the juvenile court's finding that [the juvenile] did not voluntarily consent to the blood draw," the court apparently listed the officer's reading of the implied consent admonition as a factor showing involuntariness . Id. In so doing, it chose to quote the sentence of the admonition most incompatible with the notion that its hearer could lawfully refuse consent: "You are, therefore, required to submit to the specified tests." Id. I would submit that the implication of this paragraph-in the context of an express holding that plainly stated the implied consent law no longer sufficed to establish consent-should have caused law enforcement agencies some pause.

¶ 38 At minimum, the supreme court's reasoning in Butler foreshadowed its ultimate holding in Valenzuela II : that the implied consent admonition, far from establishing voluntary consent, actually constituted an assertion of lawful authority that "effectively proclaimed that Valenzuela had no right to resist the search." Valenzuela II , 239 Ariz. 299, ¶ 22, 371 P.3d 627. Even assuming arguendo that Butler did not imply any conclusion that the admonition was overtly coercive, that case placed police agencies on plain notice that a defendant's acquiescence to the admonition would not alone establish voluntary consent under traditional Fourth Amendment standards.

¶ 39 Yet, the instant case demonstrates the Oro Valley Police Department persisted in using a procedure that relied exclusively on such acquiescence to secure consent to conduct blood draws. If it did so because it neglected to stay abreast of evolving case law, that failure would constitute "systemic negligence." See Havatone , 241 Ariz. 506, ¶¶ 21-24, 389 P.3d 1251 (good-faith exception does not apply where agency fails to consider existing precedent that casts constitutional doubt on police practice); State v. Kjolsrud , 239 Ariz. 319, ¶ 20, 371 P.3d 647 (App. 2016) (law enforcement agencies expected to "be aware of 'reasonable' interpretations of existing case law"), quoting State v. Mitchell , 234 Ariz. 410, ¶ 31, 323 P.3d 69.

¶ 40 On the other hand, any state agency that was alert to evolving precedent-and that sought to conform its search procedures to the holding in Butler -would review the content of the implied consent admonition and evaluate its suitability to establish voluntary consent. Such an inquiry could not avoid considering whether the language of the admonition was logically or semantically compatible with securing voluntary consent from a suspect.

¶ 41 Nor could any agency earnestly assess the constitutionality of its blood draw search procedure without considering Bumper , 391 U.S. at 548-50, 88 S.Ct. 1788. That case turned on the distinction between voluntary consent and submission to an officer's claim of lawful authority. Id. It unambiguously held that an officer's demand to search, given under color of legal authority, was not compatible with a state claim that a suspect had voluntarily consented. Id. Bumper could not be easily overlooked by any agency conducting a review of its practices under Butler . Bumper was the very case cited in Butler for the standard that consent must be "freely and voluntarily given." Butler , 232 Ariz. 84, ¶ 19, 302 P.3d 609.

¶ 42 For the above reasons, a serious review of police practice after Butler should have generated doubts about the voluntariness of any acquiescence to the implied consent admonition. If those doubts did not arise *455from the semantic disharmony between the words "require" and "submit" when compared to the words "voluntary" and "consent," they should have arisen upon reading longstanding, and clearly pertinent, jurisprudence from the United States Supreme Court.

¶ 43 The majority maintains, however, that "the law does not require law enforcement to make legal assessments our courts have not made." And, it asserts that "at the time of Weakland's arrest, the courts of this state had not concluded the admin per se admonition was coercive or otherwise negated consent after Butler ." It supports that conclusion by citing jurisprudence from an inferior Arizona court, issued after the time of Weakland's arrest, which found acquiescence to the admonition an act of voluntary consent. See supra ¶19 (citing Valenzuela I and Arizona Court of Appeals cases following it under principles of stare decisis).

¶ 44 In essence, the majority asks: If some reasonable appellate judges could later find no constitutional infirmity with the admonition after Butler , how could we hold the state to any higher standard? But this intuitively appealing argument overlooks the express holding of Butler , provides an incomplete assessment of the subsequent case law, and implicitly assesses good faith under an incorrect standard.

¶ 45 To be sure, police agencies need not make legal assessments our courts have not made. See Kjolsrud , 239 Ariz. 319, ¶ 20, 371 P.3d 647 (law enforcement agencies not "expected to anticipate developments in the law"), quoting Mitchell , 234 Ariz. 410, ¶ 31, 323 P.3d 69. But law enforcement does have a duty, if it later seeks to claim good faith, to adapt to our state supreme court's express holdings and the clear implications of the court's reasoning. As discussed, Butler unambiguously held that the state's implied consent law did not itself establish voluntariness-and it proceeded to find a lack of voluntariness notwithstanding Butler's acquiescence to the implied consent affidavit.

¶ 46 The majority is correct that the Butler opinion did not expressly hold that the admonition was affirmatively coercive. But Butler could be correctly cited for the proposition that the implied consent admonition was not sufficient alone to facilitate voluntary consent.

¶ 47 To the extent the majority suggests that case law post-dating Weakland's arrest should inform our assessment of the state's good faith in drawing Weakland's blood, it fails to articulate comprehensively the weight of that jurisprudence. Subsequent to the holding in Butler , only two courts published opinions, unconstrained by stare decisis, that squarely addressed whether the language of the implied consent admonition was compatible with constitutionally valid consent: this division of the Arizona Court of Appeals (in Valenzuela I ) and the Arizona Supreme Court (in Valenzuela II ). This court agreed with the state, in a divided decision, and held that it was compatible. The Arizona Supreme Court held unanimously that it was not.

¶ 48 In all, five justices and one judge who addressed the issue held that the implied consent affidavit was incompatible with securing voluntary consent. Only two judges held otherwise. That blunt head count-that most jurists ultimately rejected the state's theory-does little to assist the state's current contention that it could have previously proceeded without harboring doubts about the constitutionality of its practice.

¶ 49 The majority emphasizes two court of appeals decisions that did not address the question of whether the language of the admonition was compatible with voluntary consent. Compare Okken , 238 Ariz. 566, ¶¶ 14-25, 364 P.3d 485 (Arizona's implied consent statute or scheme , imposing negative consequences for refusing a blood test, did not render consent involuntary) and Pena, No. 2 CA-CR 2013-0241 (defendant's decision to provide consent, after previously withholding it, not coerced) with Valenzuela II , 239 Ariz. 299, ¶¶ 22-28, 371 P.3d 627 (admonition not compatible with consent because its language implies officers have authority to command acquiescence). In citing these cases, my colleagues overlook that our appellate courts do not commonly address alleged constitutional infirmities not raised or briefed by the parties on appeal. See *456Calnimptewa v. Flagstaff Police Dep't, 200 Ariz. 567, ¶ 24, 30 P.3d 634 (App. 2001) (appellate opinions should not be read as authority for matters not "specifically presented and discussed"). For that reason, neither case can be read as implicitly holding that the language of the admonition was compatible with securing voluntary consent to a blood draw. Neither court was asked to address that question.

¶ 50 Properly understood in the context of a good-faith analysis, the pertinent history of Arizona jurisprudence can be summarized as follows. In Butler , the Arizona Supreme Court clearly drew the constitutionality of the state's practice here into question-even if it did not expressly hold it coercive. After Butler , the state's use of the admonition persisted in the face of, at best, unsettled law.

¶ 51 At that time, a careful review of the admonition's language, together with an appraisal of the pertinent longstanding jurisprudence, Bumper , should have raised substantial doubts about the constitutionality of the practice. In subsequent jurisprudence, most Arizona jurists tasked with squarely addressing the question in the first instance-and all five justices of our highest court-rejected the state's claim that acquiescence to the admonition demonstrated voluntary consent. Valenzuela II , 239 Ariz. 299, ¶ 25, 371 P.3d 627 ; Valenzuela I , 237 Ariz. 307, ¶¶ 50, 57, 350 P.3d 811 (Eckerstrom, C.J., dissenting).

¶ 52 None of this jurisprudential history gives comfort to the state's claim that it could, after Butler , assume it was securing blood draws in conformity with constitutional standards. Rather, by persisting in the use of the admonition as the sole basis for consent, it disregarded a plain risk that the practice would not pass constitutional muster.

¶ 53 In this legal context, we are provided two potential standards for assessing the good-faith exception. Davis instructs that when the prevailing law at the time of the state's unconstitutional practice had "specifically authorize[d]" the practice, the good-faith exception applies unless the police officers exhibit "deliberate, reckless, or grossly negligent disregard for Fourth Amendment rights." 564 U.S. at 251, 257, 131 S.Ct. 2419. By contrast, as our supreme court clarified in Havatone , the state enjoys no such protection from the exclusionary rule when the law pertaining to the practice is "unsettled." Havatone , 241 Ariz. 506, ¶ 29, 389 P.3d 1251. Under that circumstance, when the pre-existing law suggests it is a "close" case or the constitutionality of the practice is "dubious," the exclusionary rule must apply to encourage the police to err on the side of constitutional behavior. Id.

¶ 54 After Butler , the law was unsettled. Nonetheless, the state persisted in its prior practice. In so doing, it disregarded a substantial risk that its practice violated a suspect's Fourth Amendment rights. To apply the good-faith exception under such circumstances would only encourage law enforcement to continue with dubious practices until a court finds them expressly unconstitutional. See id. (rejecting application of good-faith exception in "close" cases, when law unsettled, because otherwise police "would have little incentive to err on the side of constitutional behavior"). Under the scenario here, suppression would serve as an effective deterrent, causing law enforcement agencies to more exactingly stay abreast of developments in the law and faithfully review their procedures in accordance therewith. See Davis , 564 U.S. at 237, 131 S.Ct. 2419.

¶ 55 When we exclude or suppress evidence in a criminal case, the social costs of doing so are distressingly concrete. An individual defendant, and perhaps others similarly situated, may escape responsibility for crimes they have committed. Davis , 564 U.S. at 237, 131 S.Ct. 2419. Indeed, as here, the ill-gotten evidence often demonstrates a defendant's guilt. For this reason, we decline to impose the exclusionary rule when law enforcement violations of constitutional rights occurred in good faith. And, when we must exclude or suppress evidence, it is never a happy judicial task. See id. But it is inevitably in criminal cases that important constitutional boundaries between the state and the individual are articulated and enforced. Mapp v. Ohio , 367 U.S. 643, 655-56, 81 S.Ct. 1684, 6 L.Ed.2d 1081 (1961). I fear we abdicate our duty to guard those boundaries when we characterize law enforcement practices, continuing in the face of their dubious *457constitutionality, as good faith. I, therefore, respectfully dissent.