(dissenting). Today’s decision is an abrupt about-face from Matter of Patchogue-Medford Congress of Teachers v Board of Educ. (70 NY2d 57), where only last year this court concluded that the proposed random urine testing of probationary teachers constituted a search, that having to urinate *443under the observation of another person was at least as intrusive as a strip search, and that — despite the teachers’ diminished expectation of privacy, the public importance of unimpaired teachers, and the prevalence of drugs in schools— such searches were constitutionally prohibited in the absence of individualized reasonable suspicion. Now, the court upholds a proposed program of random urine testing under the observation of another person without any individualized suspicion, and without any reviewable procedural safeguards yet in place, because this is an elite voluntary corps of police officers with access to narcotics, because of the strong public interest in the purity and integrity of these employees, and because of their diminished privacy interest.
The criteria the court identifies are applicable to a wide array of public employees — even teachers, who are in an environment where drug use is particularly high.1 I disagree with the departure from Patchogue and with the new boundless test for permissible searches. The court should not approve this program on the present record, or in its present procedural posture. To my mind, the comprehensive trial court (133 Misc 2d 544) and Appellate Division (131 AD2d 214) opinions correctly state the law that governs this case, and I therefore respectfully dissent.
In that the proposed testing undisputedly constitutes a "search” — meaning government intrusion into petitioner’s legitimate expectation of privacy (Katz v United States, 389 US 347) — analysis begins with the governing constitutional principle that searches must not be unreasonable (NY Const, art I, § 12; US Const 4th Amend). The usual constitutional standard is not reasonable suspicion; it is that a search may be conducted only pursuant to a warrant approved by a neutral, detached Magistrate, based upon probable cause to believe that the party to be searched has violated the law.
There are a few "jealously and carefully drawn” exceptions to the warrant-preference rule (Jones v United States, 357 US 493, 499), most of them involving situations where probable *444cause exists — for instance, searches upon arrest, searches under exigent circumstances, and searches of objects in plain view. Within the category of warrantless searches there exists another group of cases in which a limited search may be conducted based upon mere reasonable suspicion that a violation has occurred (see, e.g., New Jersey v T. L. O., 469 US 325 [school search]; Delaware v Prouse, 440 US 648 [discretionary automobile stops]; United States v Brignoni-Ponce, 422 US 873 [roving-patrol automobile stops in border areas]; Terry v Ohio, 392 US 1 ["stop and frisk”]). These cases, however, represent exceptional circumstances in which special needs make the warrant and probable cause requirement impracticable. Even in these exceptional cases, individuals are not, without more, subjected to so great an intrusion on their individual privacy and dignity as providing urine samples (see, Matter of Patchogue-Medford Congress of Teachers v Board of Educ., supra, at 69; People v Scott, 63 NY2d 518, 525).
The drug-testing program before us is thus three steps removed from the warrant requirement. It does not fall within any of the carefully drawn exceptions to the warrant-preference rule involving probable cause, and it does not fall within the exceptional circumstances involving individualized reasonable suspicion. Although the court has hypothesized that searches might at times be permitted even without individualized suspicion, we have made clear that such searches will be "closely scrutinized, and generally only permitted when the privacy interests implicated are minimal, the government’s interest is substantial, and safeguards are provided to insure that the individual’s reasonable expectation of privacy is not subjected to unregulated discretion.” (Matter of Patchogue-Medford Congress of Teachers v Board of Educ., supra, at 70.)
Those prerequisites are not satisfied on this record.
Most significantly, the Commissioner has demonstrated no need in this case for surprise drug testing. The majority opinion incorrectly inverts and thus bypasses the proper order of analysis, reasoning from the premise that the OCCB officers have a reduced expectation of privacy to the conclusion that they therefore may be searched without cause. Unless at the outset the standard of reasonable suspicion is inadequate to allow identification of substance-abusing officers, the government lacks an interest that justifies departure from an individualized suspicion standard, and a random search is impermissible (see, 3 LaFave, Search & Seizure § 10.3 [e], at 47-48 [1988 Pocket Part]).
*445The Police Commissioner’s own statistics show 10 instances where narcotics-related disciplinary charges were filed against OCCB members over a four-year period. As the majority recognizes, this is "meager and unpersuasive to support an assertion of pervasive illegal drug use”. (Majority opn, at 442.) The Commissioner does not suggest that job performance declined or unusual incidents increased in a way consistent with pervasive undetected drug use. There is nothing to indicate that the prevalence of drug use in the general public is also a problem among OCCB officers. Indeed, the only evidence before us is decidedly to the contrary, supporting the conclusion that with the heightened scrutiny to which they are already subjected, "those finally accepted for membership in OCCB are among the best in the force and of a nature even less likely to use drugs than others in the Department.” (133 Misc 2d 544, 552.)
The important goal of deterrence — to the extent that it is suggested as the object of this program — is not, without evidence of a pervasive problem among the persons to be tested, justification for this search. Many types of illegal conduct could be deterred were the State permitted to search everyone "periodically in an all-inclusive dragnet * * * By restricting the government to reasonable searches, the State and Federal Constitutions recognize that there comes a point at which searches intended to serve the public interest, however effective, may themselves undermine the public’s interest in maintaining the privacy, dignity and security of its members.” (Matter of Patchogue-Medford Congress of Teachers v Board of Educ., supra, at 70.)
The majority identifies as a strong government interest the integrity of the highly specialized OCCB unit itself. There are references to elite, special, highly sensitive employment. The implicit conclusion is that certain public employees, simply by virtue of their position and without any proof of wrongdoing, may be required to demonstrate their innocence upon demand. This not only provides little guidance as to who is subject to random tests, but also reduces the constitutional limitations on the government’s power to search to a hollow promise, and cannot be squared with Patchogue. If the integrity of a group of public employees is enough of an interest to justify random searches of ostensibly innocent persons it is difficult to imagine who should be exempt from such intrusion. The public clearly has an interest in the integrity of every *446employee who serves it; equally clearly, that alone does not give the State the power to conduct intrusive searches of their person.
The cases cited by the majority do not support its conclusion. Overwhelmingly courts have concluded — as we did in Patchogue — that a reasonable suspicion standard adequately protects the public interest.2
Nor does Professor LaFave’s comment (majority opn, at 438) support the majority’s conclusion. After a lengthy review of cases overturning random testing programs, Professor LaFave wrote: "[I]t must be acknowledged that there may be a very few limited circumstances in which the general line of analysis pursued above will not produce the conclusion that the reasonable suspicion standard strikes the appropriate balance. Perhaps there are a few forms of public employment in which the hazards of even a momentary lapse of attention or judgment are so substantial in terms of the physical danger to fellow workers or the general public, and in which the opportunities for preventive close supervision are so limited that only random or blanket testing will suffice” (3 LaFave, Search & Seizure § 10.3 [e], at 47-48 [1988 Pocket Part]).
The skimpy record in this case does not support the proposition that OCCB members are an example of Professor La-*447Fave’s very limited theoretical exception. The factual submissions before us show only that the routine duties of OCCB members include enforcement of the narcotics laws, and that these officers have ready access to drugs. The rest — including graphic references to drug traffickers in a nether world, covert exchanges of classified information, and unique safety and other hazards of high-risk assignments — is solely lawyers’ rhetoric in briefs with no substantiation in the record. Perhaps a better factual record of what OCCB members do — if in actuality they do fit Professor LaFave’s theoretical exception —would have permitted the courts to fashion standards that could then serve as guidance in this important, developing area of law. We don’t have that record. There has been no showing that the heightened scrutiny to which OCCB members are already subjected does not, and would not — together with the reasonable suspicion standard — fully protect the public interest.
Thus, I would affirm the Appellate Division order simply because this record does not contain facts establishing the threshold requirement of government need for random searches.
While placing these officers "in fishbowl-like circumstances undreamed of by Calpurnia herself’ (majority opn, at 440), the majority nonetheless says that OCCB members retain important personal privacy rights, particularly the right to be secure against unreasonable searches and seizures of their persons, homes, cars, lockers and other personal effects (majority opn, at 442). I believe these retained privacy rights include the right to be free of the "very serious,” "severe privacy intrusion” (majority opn, at 438) now proposed.
The court places great emphasis on the "very diminished” or "minimal”3 privacy expectations of OCCB members centering on the voluntary nature of the assignment, the several drug tests to which they are routinely subjected and the scrutiny given their personal and financial affairs. While again these generalized assertions of heightened scrutiny are *448not supported more particularly in the factual record, the intensified examination these officers already receive — together with the documented lower incidence of drug abuse found among them, and the reasonable suspicion standard— point to the opposite conclusion from that reached by the court: that there is no justification for random testing of these individuals.
As described by the majority, the diminished or minimal privacy theory appears to proceed on a sort of implied consent —that voluntary pursuit of service in an elite corps where standardized drug testing is known in advance implies acceptance of further testing. Notably, in Patchogue we observed that the urine testing required of all probationary school teachers was "accepted and traditional” (70 NY2d, at 70, supra), that they had a "diminished expectation of privacy with respect to State inquiries into their physical fitness to perform as teachers, and it is not unreasonable to require teachers to submit to further testing when school authorities have reason to suspect that they are currently unfit for teaching duties.” (Matter of Patchogue-Medford Congress of Teachers v Board of Educ., supra, at 69.) That same rationale should apply here. Knowledge of standardized drug testing cannot itself be deemed consent to additional testing, that in effect then becomes consent to more intrusive searches (see, Lo-Ji Sales v New York, 442 US 319, 329).
Finally, in Patchogue we made crystal-clear that any random search program would be permitted only where there were adequate safeguards provided to insure that the individual’s reasonable expectation of privacy was not subjected to unregulated discretion. Now the court for the first time approves a program for random searches without reasonable suspicion — thus establishing a precedent — yet it does so in a case where there are as yet no reviewable safeguards.4
In Patchogue the court recognized that "there comes a point at which searches intended to serve the public interest, however effective, may themselves undermine thé public’s interest in maintaining the privacy, dignity and security of its mem*449bers.” (Matter of Patchogue-Medford Congress of Teachers v Board of Educ., supra, at 70.) We made clear that random searches without reasonable suspicion would be closely scrutinized and permitted only upon adherence to specified requirements. Today we reaffirm those sentiments in words. In deed, however, the court gives its imprimatur to a program not yet even fully formulated, on factual submissions that do not satisfy the legal thresholds we only recently identified. On this record I would affirm the Appellate Division order.
Chief Judge Wachtler and Judges Simons, Alexander and Hancock, Jr., concur with Judge Bellacosa; Judge Kaye dissents and votes to affirm in a separate opinion in which Judge Titone concurs.
Order reversed, etc.
. The majority points to distinctions between teachers and these police officers. Of course there are differences in the nature of their work. Yet in certain central respects the analogy is strong. Those who voluntarily choose to become teachers of the young should particularly be people of integrity, unimpaired by and uninvolved in drugs. The Police Commissioner’s statistics indicáte approximately 60% of students in grades 7-12 have had some involvement with illicit drugs; one of every two students reported some drug use.
. Even as to police officers, courts overwhelmingly have required reasonable suspicion. (See, e.g., Penny v Kennedy, 846 F2d 1563 [6th Cir 1988] [mandatory urinalysis of police officers on department-wide basis without reasonable suspicion constitutes an unreasonable search]; Feliciano v City of Cleveland, 661 F Supp 578, 589 [ND Ohio 1987] ["reasonable individualized suspicion that a police officer is using illicit drugs must be required for urinalysis to be reasonable”]; American Fedn. of Govt. Employees v Weinberger, 651 F Supp 726 [SD Ga 1986] [Federal police officers cannot be tested without reasonable suspicion]; Bostic v McClendon, 650 F Supp 245 [ND Ga 1986] [mass testing of urine of police officers without reasonable suspicion violates Fourth Amendment]; Capua v City of Plainfield, 643 F Supp 1507 [D NJ 1986] [random urine testing of members of police department in absence of reasonable suspicion violates the Fourth Amendment]; Fraternal Order of Police v City of Newark, 216 NJ Super 461, 524 A2d 430 [1987] [city directive requesting all members of narcotics unit to submit to urine testing without probable cause or reasonable suspicion violates State and Federal Constitutions]; Turner v Fraternal Order of Police, 500 A2d 1005, 1009 [DC App 1985] ["There must be a reasonable objective basis to suspect that a urinalysis will produce evidence of an illegal drug use”]; City of Palm Bay v Bauman, 475 So 2d 1322 [Fla App 1985] [police officers and firefighters cannot be tested without reasonable suspicion]; see also, Security & Law Enforcement Employees v Carey, 737 F2d 187 [2d Cir 1984] [holding unconstitutional random strip and cavity searches of prison employees for contraband].)
. The notion of "minimal” privacy expectations of OCCB members is curious. We all agree that the proposed test constitutes a "search,” which is by definition an intrusion into an individual’s legitimate expectation of privacy. Having concluded that there is an intrusion into the legitimate expectation of privacy of these officers, the test of reasonableness requires consideration of "the scope of the particular intrusion, the manner in which it is conducted, the justification for initiating it, and the place in which it is conducted.” (Bell v Wolfish, 441 US 520, 559.)
. What the court refers to as "facial” validity (majority opn, at 435, 441, 442) is a novel — and I believe incorrect — application of the concept of "facial” and "as applied” constitutional challenges. The review here is termed "facial” only because the program has not yet been fully formulated — indeed added procedural safeguards have been suggested even in the briefs before this court. If anything, incompleteness of the proposed program would point to lack of ripeness or even to facial invalidity, but not to facial validity.