Justice Kagan, with whom Justice Sotomayor and Justice Jackson join, dissenting.

No one doubts that the distribution of sexually explicit speech to children, of the sort involved here, can cause great harm. Or to say the same thing in legal terms, no one doubts that States have a compelling interest in shielding children from speech of that kind. What is more, children have no constitutional right to view it. The Texas statute before us (H. B. 1181) addresses speech understood in First Amendment law as “obscene for minors.” That label means the First Amendment does not protect the speech for minors. The State can restrict their access without fear of colliding with the Constitution.

The trouble comes in the last two sentences’ italics. Speech that is obscene for minors is often not so for adults. For them, the category of obscene—and therefore unprotected speech—is narrower. See ante, at 8–10. So adults have a constitutional right to view the very same speech that a State may prohibit for children. And it is a fact of life—and also of law—that adults and children do not live in hermetically sealed boxes. In preventing children from gaining access to “obscene for children” speech, States sometimes take measures impeding adults from viewing it too—even though, for adults, it is constitutionally protected expression. What, then, to do?

Cases raising that question have reached this Court on no fewer than four prior occasions—and we have given the same answer, consistent with general free speech principles, each and every time. Under those principles, we apply strict scrutiny, a highly rigorous but not fatal form of constitutional review, to laws regulating protected speech based on its content. See ante, at 6. And laws like H. B. 1181 fit that description: They impede adults from viewing a class of speech protected for them (even though not for children) and defined by its content. So when we have confronted those laws before, we have always asked the strict-scrutiny question: Is the law the least restrictive means of achieving a compelling state interest? See ibid. There is no reason to change course.

A law like H. B. 1181 might well pass the strict-scrutiny test, hard as it usually is to do so. As just noted, everyone agrees that shielding children from exposure to the sexually explicit speech H. B. 1181 targets is a compelling state interest. And Texas might be right in arguing that it has no less restrictive way to achieve that goal: It is difficult, as everyone also agrees, to limit minors’ access to things appearing on the internet. If H. B. 1181 is the best Texas can do—meaning, the means of achieving the State’s objective while restricting adults’ speech rights the least—then the statute should pass First Amendment review.

But what if Texas could do better—what if Texas could achieve its interest without so interfering with adults’ constitutionally protected rights in viewing the speech H. B. 1181 covers? That is the ultimate question on which the Court and I disagree. The majority says that Texas may enforce its statute regardless, because only intermediate scrutiny applies and that test does not ask whether a State has adopted the least speech-restrictive means available. I disagree, based on conventional First Amendment rules and the way we have consistently applied them in this very context. The State should be foreclosed from restricting adults’ access to protected speech if that is not in fact necessary.

The majority’s opinion concluding to the contrary is, to be frank, confused. The opinion, to start with, is at war with itself. Parts suggest that the First Amendment plays no role here—that because Texas’s law works through age verification mandates, the First Amendment is beside the point. See ante, at 13–18. But even the majority eventually gives up that ghost. As, really, it must. H. B. 1181’s requirements interfere with—or, in First Amendment jargon, burden—the access adults have to protected speech: Some individuals will forgo that speech because of the need to identify themselves to a website (and maybe, from there, to the world) as a consumer of sexually explicit expression. But still, the majority proposes, that burden demands only intermediate scrutiny because it arises from an “incidental” restriction, given that Texas’s statute uses age verification to prevent minors from viewing the speech. See ante, at 13, 18–19. Except that is wrong—nothing like what we have ever understood as an incidental restraint for First Amendment purposes. Texas’s law defines speech by content and tells people entitled to view that speech that they must incur a cost to do so. That is, under our First Amendment law, a direct (not incidental) regulation of speech based on its content—which demands strict scrutiny.

The majority’s attempt to distinguish our four precedents saying just that rounds out the list of its errors. According to the majority, all of those decisions involved prohibiting rather than merely burdening adults’ access to obscene-for-children speech. See ante, at 21. But that is not true. And in any event it would not matter: The First Amendment prevents making speech hard, as well as banning it outright. So on all accounts the majority’s rationale craters.

The majority is not shy about why it has adopted these special-for-the-occasion, difficult-to-decipher rules. It thinks they are needed to get to what it considers the right result: giving Texas permission to enforce its statute. See ante, at 19–21. But Texas should not receive that permission if it can achieve its goal as to minors while interfering less with the speech choices of adults. And if it cannot, then Texas’s statute would survive strict scrutiny, given the obvious importance of its goal. For that reason, the majority’s analysis is as unnecessary as it is unfaithful to the law.

I

Under ordinary First Amendment doctrine, this Court should subject H. B. 1181 to strict scrutiny. That is because H. B. 1181 covers speech constitutionally protected for adults; impedes adults’ ability to view that speech; and imposes that burden based on the speech’s content. Case closed. And making the right answer yet more obvious, we have said as much four times before, when reviewing statutes imposing similar content-based burdens on protected sexually explicit speech. So the case is closed even tighter: The standard should be strict scrutiny. The only open question here should be whether H. B. 1181 can satisfy that test.

A

No one (not even Texas, not even the majority) disputes that H. B. 1181 covers a substantial amount of speech protected by the First Amendment. We have, of course, often held that obscene speech, as defined in Miller v. California, 413 U. S. 15 (1973), is not so protected. But H. B. 1181 does not use the ordinary Miller test (relating to prurience, offensiveness, and value) as the trigger for regulation. Instead, it adapts each part of that test “for minors,” thus covering speech that is “obscene from a child’s perspective.” Tex. Civ. Prac. & Rem. Code Ann. §129B.002(a) (West Cum. Supp. 2024); ante, at 9 (emphasis deleted). And that child-centric category of speech extends wider than the traditional obscenity category. See ante, at 9–10. In the gap between the two is much sexually explicit speech that adults have every right to view. For adults cannot be limited to “only what is fit for children.” Butler v. Michigan, 352 U. S. 380, 383 (1957). Their right to view “[s]exual expression,” outside the traditional obscenity category, is “protected by the First Amendment.” Sable Communications of Cal., Inc. v. FCC, 492 U. S. 115, 126 (1989).

And H. B. 1181 impedes the exercise of that right. Recall how the statute works. To enter a covered website—with all the protected speech just described—an individual must verify his age by using either a “government-issued identification” like a driver’s license or “transactional data” associated with things like a job or mortgage. §§129B.001(7), 129B.003(b)(2); see ante, at 2–3. For the would-be consumer of sexually explicit materials, that requirement is a deterrent: It imposes what our First Amendment decisions often call a “chilling effect.” E.g., Americans for Prosperity Foundation v. Bonta, 594 U. S. 595, 606 (2021). It is not, contra the majority, like having to flash ID to enter a club. See ante, at 14–15. It is turning over information about yourself and your viewing habits—respecting speech many find repulsive—to a website operator, and then to  \. \. \. who knows? The operator might sell the information; the operator might be hacked or subpoenaed. We recognized the problem in a case involving sexual material on cable TV: Similar demands, we decided, would “restrict viewing by subscribers who fear for their reputations should the operator, advertently or inadvertently, disclose the list of those who wish to watch the ‘patently offensive’ channel.” Denver Area Ed. Telecommunications Consortium, Inc. v. FCC, 518 U. S. 727, 754 (1996). The internet context can only increase the fear. And the Texas law imposes costs not just on potential users, but on website operators too. They must either implement a system costing (the District Court found) at least $40,000 for every 100,000 verifications, or else pay penalties of $10,000 per day. See §129B.006(b); Free Speech Coalition, Inc. v. Colmenero, 689 F. Supp. 3d 373, 385–386 (WD Tex. 2023). Those expenses, Texas boasts, have already caused one major operator to exit the State’s market. See Brief for Respondent 21. So in multiple ways, H. B. 1181 burdens expression.

Finally, H. B. 1181 imposes those burdens on protected speech based on the speech’s “communicative content,” making it a quintessential content-based law. Reed v. Town of Gilbert, 576 U. S. 155, 163 (2015). A statute, we have often said, is content-based on its face when it “draws distinctions” based on the “topic,” “subject matter,” “idea,” or “message expressed.” E.g., ibid. H. B. 1181 does just that. It applies when more than a third of the expression on a website is “sexual material” of a certain kind (prurient, offensive, and valueless for minors). §§129B.001(6), 129B.002(a). And whether expression qualifies as such material depends entirely on what it depicts. If the website has the requisite sexually explicit content, the regulation kicks in. Alternatively, if that content is absent (if, say, the website focuses on politics or sports), the regulation does not. “That is about as content-based as it gets.” Barr v. American Assn. of Political Consultants, Inc., 591 U. S. 610, 619 (2020) (plurality opinion). Not even the majority disputes the point. See ante, at 28.

All of that leads, under well-settled law, to just one conclusion: H. B. 1181 is subject to strict scrutiny. Take a law burdening protected speech based on its content—as H. B. 1181 does for every adult—and the standard of review follows in its wake. See Turner Broadcasting System, Inc. v. FCC, 512 U. S. 622, 642 (1994) (We “apply the most exacting scrutiny to regulations that suppress, disadvantage, or impose differential burdens upon speech because of its content”). It does not matter whether we are persuaded in a given case that the State, in passing the regulation, had laudable, even compelling aims. See Reed, 576 U. S., at 166 (A good “justification cannot transform a facially content- based law into” the opposite). Those interests are considered only in applying strict scrutiny, not in deciding whether that is the right standard to be applied. Over the years, we have recited the governing rule almost like a mantra: If a law burdens protected speech based on what that speech says or depicts—as H. B. 1181 does—the law has to clear the strict-scrutiny bar.

B

What is more, our precedents have applied that rule in four cases similar to this one—when a statute has limited adults’ access to sexually explicit materials in order to prevent those materials from getting to minors. The laws at issue pertained to diverse media—the telephone, cable, and (twice, as here) the internet. But the analysis about the level of scrutiny was in each case the same. To show the Court’s (previous) consistency—and its relevance today—it is worth reviewing them one by one by one by one.

In Sable Communications v. FCC, the Court considered a statute directed at dial-a-porn services that prohibited sexually “indecent” telephone messages, extending beyond those obscene for adults under Miller. 492 U. S., at 122–123. The Government defended the law as an effort to protect children from exposure to the speech. See id., at 128. We recognized that interest as compelling. See id., at 126. But we also understood that adults had a “protected” First Amendment right to listen to the non-obscene indecent speech that the law covered. Ibid. And so the Court applied strict scrutiny—thus requiring the Government to show that the statute did not “unnecessarily interfer[e] with [adults’] First Amendment freedoms.” Ibid.

Then, in Reno v. American Civil Liberties Union, 521 U. S. 844, 859–861 (1997), the Court addressed a statute barring internet transmissions of obscene, indecent, or “patently offensive” messages to those under 18, with an affirmative defense available to anyone making use of age verification measures. Although the statute encompassed only communications to minors and excused from penalties those using a “reasonable” method to verify age, the Court recognized the “burden” that the statute would impose “on adult speech.” Id., at 860, 874. Because of that “interfere[nce] with adult-to-adult communication”—and despite the significance of the Government interest “in protecting children”—the Court again insisted on applying strict scrutiny. Id., at 875–876. So once more the key issue was whether “less restrictive alternatives would be at least as effective in achieving” the Government’s goals. Id., at 874.

Next, in United States v. Playboy Entertainment Group, Inc., 529 U. S. 803, 806 (2000), the Court evaluated a law requiring that “sexually-oriented” cable channels “limit their transmission to hours when children are unlikely to be viewing.” “[W]hat standard [must] the Government” meet, the Court asked, for the law to survive? Id., at 814. We did not think the question close. “As we consider a content-based regulation” of “protected speech,” we said, “the answer should be clear: The standard is strict scrutiny.” Ibid.; see id., at 812–813. So “if a less restrictive means” would serve the Government’s goals, “the Government must use it.” Id., at 815. Otherwise, the Court explained, the Government could, contrary to the First Amendment, “restrict speech without an adequate justification.” Id., at 813.

And the denouement: The statute the Court addressed in Ashcroft v. American Civil Liberties Union, 542 U. S. 656 (2004), was a near-twin of Texas’s. The Child Online Protection Act (COPA) prohibited commercial entities from posting on the internet content “harmful to minors.” Id., at 661 (quoting 47 U. S. C. §231(a)(1)). And just like H. B. 1181, that statute defined the covered material by adapting the Miller obscenity test for children—thus creating a category of obscene-for-children speech. See 542 U. S., at 661–662; supra, at 4. So too, COPA made the adoption of an age verification system crucial. It did so by providing an affirmative defense to any entity that verified age through an “adult personal identification number” or similar mechanism before granting access to the posted materials. Ashcroft, 542 U. S., at 662. So, as in H. B. 1181, if the poster verified age, no liability could attach. How, then, to analyze such a statute? The Court viewed the problem as it had in prior cases: COPA, though directed at keeping sexually explicit materials from children, “was likely to burden some speech that is protected for adults.” Id., at 665. And because of that “content-based restriction[],” the Court needed to apply strict scrutiny. Id., at 660, 665, 670. The Government thus had to show that “the proposed alternatives will not be as effective as the challenged statute.” Id., at 665. In short, Ashcroft adhered to the view that “ ‘the governmental interest in protecting children from harmful materials’ does not ‘justify an unnecessarily broad suppression of speech addressed to adults.’ ” Lorillard Tobacco Co. v. Reilly, 533 U. S. 525, 581 (2001) (Thomas, J., concurring in part and concurring in judgment) (quoting Reno, 521 U. S., at 875).¹

Four times, one result. Which is not surprising, because it is the result that basic First Amendment principles command. A statute tries to cut off children’s access to sexually explicit speech, in line with the most worthy objectives. But the statute as well impedes adults’ access to that speech, which the First Amendment protects. And the statute does so by drawing content-based lines: Sexually explicit speech is burdened, other speech is not. It follows, as the night the day, that strict scrutiny applies—that the statute, in addition to serving a compelling purpose, can restrict only as much adult speech as is needed to achieve the State’s goal. That is true in the four cases above, and it is true in this case too.

C

Applying strict scrutiny in this context, however, need not be a death sentence. To the contrary, a State exercising care should be able to devise a regulatory means of achieving its objective consistent with the First Amendment.

The first part of the strict-scrutiny test is here easy to meet. The majority is right that a State has a compelling interest in shielding children from the obscene-for-children materials that H. B. 1181 covers. See ante, at 32. This Court has said as much before. See Sable, 492 U. S., at 126 (recognizing a “compelling interest in protecting the physical and psychological well-being of minors,” which “extends to shielding” them from materials “not obscene by adult standards”); Denver Area, 518 U. S., at 743 (plurality opinion) (noting that the interest in “protect[ing] children from exposure to patently offensive sex-related material” is “one that this Court has often found compelling”). And a State is entitled to think that the need has become only more urgent over the years, given the time children now spend online and the materials they can find there. See ante, at 27.

The critical question, then, is whether the State can show that it has limited no more adult speech than is necessary to achieve its goal. Or said another way (in fact, Ashcroft’s way), whether the State can show that “the proposed alternatives will not be as effective as the challenged statute.” 542 U. S., at 665. If the State cannot, the statute should not take effect, because it would limit protected speech unnecessarily. There would be every reason to make the State switch to a less-speech-restrictive, equally-or-more-effective regulatory mechanism. But a State that has closely attended to the speech consequences of its regulation might well make the required showing in this sphere. Given how the internet works, no court should expect that a law effectively shielding children from sexually explicit expression could leave adults wholly unaffected. To the contrary, such a law will almost necessarily impose corollary burdens. And Texas may be right that the commonly proposed alternatives to laws like H. B. 1181—such as content filtering technology—cannot equal, or even approach, age verification systems in effectiveness. See Brief for Respondent 37–38. In that event, those alternatives will be irrelevant to the inquiry, and a court will explore only whether another, equally effective age verification mechanism will place a lesser burden on protected speech. Review of that kind should not be the horror show for Texas and other States that the majority maintains. See ante, at 19–21. It is just what they should have to pass before implementing a content-based burden on protected expression.

II

How does the majority reach a different result?

The analytic path of today’s opinion is winding, but I take the majority to begin with a conviction about where it must not end—with strict scrutiny. The majority is not so coy about this backwards reasoning. To the contrary, it defends it. See ante, at 30. The “legitimacy” of age verification schemes for sexually explicit speech, the majority tells us, is “uncontroversial” (despite Reno and Ashcroft). Ante, at 30. And “[a]pplying the more demanding strict-scrutiny standard would call” those schemes “into question.” Ante, at 19. Ergo, its conclusion. I have just explained why the majority’s fear is overblown—why in fact carefully drawn age verification laws stand a real chance of surviving strict scrutiny. But suppose I am wrong. Suppose there are both less speech-restrictive and equally effective ways to accomplish the State’s goal of protecting children from sexually explicit materials. In that event, strict scrutiny tells us, the State should use those constitutionally superior alternatives. And why argue with that? The usual way constitutional review works is to figure out the right standard (here, strict scrutiny because H. B. 1181 is content-based), and let that standard work to a conclusion. It is not to assume the conclusion (approve H. B. 1181 and similar age verification laws) and pick the standard sure to arrive there. But that is what the majority does. To answer what standard of scrutiny applies, the majority first spends four pages lauding age verification schemes as “common,” “traditional,” “appropriate,” and “necessary.” Ante, at 13–18. In other words, all over the place, and a good thing too. No wonder the majority doesn’t land on strict scrutiny.

The more puzzling question is how the majority’s reasoning fits with the idea that the First Amendment plays any role at all. For quite some time in today’s opinion, speech rights are pushed to the sidelines, or entirely off the field. Age verification schemes are just age verification schemes—again, “common,” “traditional,” “appropriate,” and “necessary.” Ibid. States use them to regulate purchases of liquor and lottery tickets and fireworks. And so, the majority says, States can also use them to regulate access to speech that is obscene for children. The power to prevent minors from gaining access to that speech “necessarily includes” the power to require proof of age. Ante, at 13. And that means, the majority concludes, that “accessing material obscene to minors without verifying one’s age is not constitutionally protected,” even for adults. Ante, at 19 (emphasis deleted). It would seem the analysis is complete. If the First Amendment does not protect adults in viewing obscene-for-children materials unimpeded by age verification, as the majority argues, then how could there be any constitutional objection to age verification laws like H. B. 1181? Or said otherwise, why would those laws have to satisfy any heightened constitutional standard, whether strict or intermediate? We have apparently arrived at a place where States can act free of all constitutional scrutiny.

But that cannot be, for reasons that by now should sound familiar. As discussed earlier, speech that is obscene for children is often not obscene for adults. See supra, at 4–5. When that is so, the First Amendment protects adults’ access to obscene-for-children speech (unlike to liquor, lottery tickets, or fireworks). Or otherwise said, the First Amendment gives them a right in that expression. And because of that protected right, different rules apply. Without a special justification, a State cannot prohibit, tax, impede, or otherwise burden an adult’s access to obscene-for-children speech. And an age verification requirement is a kind of burden. It may be smaller or larger—compare flashing ID in a store with (in the majority’s own example) having to produce “an affidavit from [a] biological parent.” Ante, at 31. It may be a simple inconvenience or it may, as suggested earlier, prevent individuals from exercising the right. See supra, at 5–6. And those differences may well matter to the conclusion when a court gets around to applying the appropriate constitutional standard. But regardless, an age verification mandate burdens an adult’s First Amendment protected right in viewing obscene-for-children expression. So a State’s power to prohibit that speech for minors does not “necessarily include[],” as the majority contends, the power to mandate age verification. Ante, at 13. It might or might not, depending on whether the mandate satisfies the constitutional scrutiny that its burden on protected speech requires.

And in the end, the majority has to accept some version of that argument. For page upon page, the majority explains that the First Amendment has nothing to say about age verification schemes attached to obscene-for-children speech. See ante, at 13–18. Again, that speech may as well be liquor, lottery tickets, or fireworks, for all it matters to the “States’ authority.” Ante, at 13. And then, in the space of one brief paragraph, the idea falls apart. Yes, the majority at last concedes, “[a]dults have the right to access speech that is obscene only to minors.” Ante, at 18. And yes, the majority admits, “submitting to age verification is a burden on the exercise of that right.” Ibid. So sure, the majority acknowledges, a really onerous age verification scheme—like its parental affidavit requirement—would flunk constitutional review. See ante, at 31. And so too, the majority says, even the least onerous mandate, like the one in Ginsberg v. New York, 390 U. S. 629, 633, 643–644 (1968), to show ID in a store, has to satisfy some form of heightened constitutional scrutiny. See ante, at 22, 31–32.2 There is no getting around the fact: Obscene-for-children speech is constitutionally protected speech for adults. See ante, at 18. And age verification schemes “burden[ing]” adults’ “right[s] to access [that] speech” are in fact not the kind of everyday, “appropriate,” and “necessary” regulation courts can wave on by. Ante, at 13–18. The Constitution, contrary to what the majority at first assured us, is now very much in the picture.

At that point, one might think, the right approach—as the Court once said—“should be clear: The standard is strict scrutiny.” Playboy, 529 U. S., at 814. Forgive a brief recap. H. B. 1181 regulates the communicative content of websites, imposing an age verification mandate on those exhibiting a specified amount of sexually explicit speech that, while obscene for children, is protected for adults. So the law directly burdens adults’ right to view speech based on its sexual content. As the Court four times before found, that means strict scrutiny applies—even though the State is attempting to prevent the speech from reaching minors. See supra, at 4–10.

The majority tries to escape that conclusion with a maneuver found nowhere in the world of First Amendment doctrine. It turns out, the majority says, that the First Amendment only “partially protects” the speech in question: The “speech is unprotected to the extent the State seeks only to verify age.” Ante, at 18, 29, n. 12 (emphasis deleted); see ante, at 28 (the speech is “unprotected to the extent that the State imposes only an age-verification requirement”). Meaning, the speech is unprotected to the extent that the State is imposing the very burden under review. Or said another way, the right of adults to view the speech has the burden of age verification built right in. That is convenient, if altogether circular. In the end, the majority’s analysis reduces to this: Requiring age verification does not directly burden adults’ speech rights because adults have no right to be free from the burden of age verification. Gerrymander the right to incorporate the burden, and the critical conclusion follows. If only other First Amendment cases were so easy!

Still, the majority must make one more move to square the circle of all it has said. Recall that notwithstanding the above, the majority has conceded that “[a]dults have the right to access” obscene-for-children speech and age verification schemes are “a burden on the exercise of that right.” Ante, at 18; see supra, at 14–15. To account for that concession in its analysis—and yet avoid strict scrutiny, as it wishes—the majority relies on a well-known distinction in First Amendment law between direct and incidental restrictions on speech. See (sorry)

E. Kagan, Private Speech, Public Purpose: The Role of Governmental Motive in First Amendment Doctrine, 63 U. Chi. L. Rev. 413, 491–505 (1996). Says the majority: The “burden experienced by adults” as a result of H. B. 1181 is “only incidental to the statute’s regulation of activity that is not protected by the First Amendment.” Ante, at 18. Or more fully (prepare for a mouthful): “The law directly regulates unprotected activity (accessing material that is obscene to minors without submitting to age verification) while only incidentally burdening protected activity (ultimately accessing that material).” Ante, at 28 (emphasis in original). And because the burden imposed on adults’ right to access the materials is only incidental, the majority concludes, only intermediate scrutiny need apply. To back up that view, the majority relies (exclusively) on United States v. O’Brien, 391 U. S. 367 (1968). See ante, at 18–19, 28.

O’Brien actually seems a good place to start in explaining why H. B. 1181 is not an incidental restriction under our law. In that case, a war protester who burned his draft card was charged with violating a statute that made it a crime for anyone to “knowingly destroy[],” “mutilate[],” or “change[]” draft registration documents. 391 U. S., at 370 (emphasis deleted). The Court assumed that O’Brien himself had engaged in expressive conduct: By burning his draft card on the steps of a government building, he was communicating opposition to the Vietnam War. See id., at 369–370, 376. But the law O’Brien broke was not about speech; it was about conduct. That law, the Court explained, prohibited all alterations of draft cards, indifferent to whether they were “public [or] private,” expressive or non-expressive. Id., at 375. So the “limitation[] on [O’Brien’s] First Amendment freedoms” was purely “incidental.” Id., at 376. And because that was so—because the statute at issue addressed only the “noncommunicative aspect” of what O’Brien did—the Court decided to apply intermediate scrutiny. Id., at 381–382.

In the years since, this Court has used the O’Brien view of incidental restrictions in several analytically identical cases—when a limitation on conduct, “having no connection with speech,” happens to sweep in a person’s expressive act. Id., at 375. In one, the National Park Service invoked a regulation that banned camping (defined to include sleeping) in designated parks to prevent a sleep-in demonstration about homelessness. Clark v. Community for Creative Non-Violence, 468 U. S. 288, 289–291 (1984). The Court applied the O’Brien standard and approved the Park Service’s action. In another, a law against “reenter[ing] a military base after having been barred by the commanding officer” was used to charge a person who had reentered a base to participate in a political demonstration. United States v. Albertini, 472 U. S. 675, 677–678 (1985). We upheld the conviction under O’Brien. And in a third, the question was whether a law banning public nudity could be applied to an establishment featuring “expressive” nude dancing. Erie v. Pap’s A. M., 529 U. S. 277, 289 (2000) (plurality opinion). Once again, we understood the case as in the O’Brien line, because a prohibition of conduct had an “incidental” effect on an expressive act. 529 U. S., at 294–295, 301. So we used the O’Brien standard, and approved the nudity ban’s application.

None of this has any bearing on H. B. 1181. That statute is not a regulation of conduct that just so happens, on occasion, to impinge on expressive activity. It is instead a direct regulation of speech, triggered by the amount of sexually explicit expression on a commercial website. Or said a bit differently: Rather than address the “noncommunicative” aspects of an activity—as all the laws described above did—H. B. 1181 regulates (and regulates only) what no one here disputes are communicative messages. O’Brien, 391 U. S., at 382. Consider: a law about altering draft certificates; a law about sleeping in parks; a law about reentering military bases; a law about public nudity; a law about sexually explicit postings on websites. Which one of those laws is not like the others? As to the first four laws, the regulation is of conduct, and the burden on expression a rare knock-on effect. As to the fifth, the regulation is of speech, and the burden on that speech the very thing the statute does. See Holder v. Humanitarian Law Project, 561 U. S. 1, 27 (2010) (noting that O’Brien applies only when the “thing actually at issue” is “conduct”). The burden H. B. 1181 imposes, of course, raises constitutional concerns only for adults. But that fact does not make the law any less a direct, not incidental, restriction on protected expression. H. B. 1181 targets communicative content, and that alone—restricting adults’ access to speech because of what it portrays, rather than because of any non-communicative element that it possesses.

And this Court, in four prior cases involving similar regulations enacted for similar reasons, has not once proposed an analogy to O’Brien. Forgive another brief recap. See supra, at 7–10. In all of those cases, States burdened protected speech for adults as a way of cutting off children’s access to the expression. Two of those efforts involved internet speech. The same two made liability for infractions turn on whether the publisher of the speech used an age verification measure. One of them—Ashcroft—defined the regulated speech identically to H. B. 1181 (using the Miller test adapted for minors). Yet in none of the four cases did even a single Justice float the idea that, because the restriction was geared toward protecting minors or involved age verification, the statute somehow effected only an incidental restriction. In every one, it was common ground (even among the dissenting Justices) that the statute’s restriction on adults’ access to speech was direct. So our precedents stand as an embarrassment to the majority’s reasoning.

The majority’s primary—and deficient—response is that those cases involved “outright bans” on speech, whereas this one involves only a burden. Ante, at 25; see ante, at 21–25. To begin with, that assertion is factually inaccurate as to three of the four. In Playboy, the law did not ban adult cable channels, but instead limited their transmission to hours when children were unlikely to be in the audience. See 529 U. S., at 806, 812. (The allowable hours were 10 p.m. to 6 a.m.—when, the District Court found, between 50% and 70% of adult viewing occurs anyway. See ibid.) So as the Court took care to explain, the statute did “not impose a complete prohibition.” Id., at 812. Rather, it effected only a “content-based burden[ ]”—as H. B. 1181 does. Ibid.3 The same is true of the statutes in Reno and Ashcroft, and in a way even more similar to Texas’s law. Recall that under those statutes, publishers using age verification measures had an affirmative defense to all liability. See supra, at 7–9. So as long as those measures were in place, publishers could confidently press send on whatever sexual content they wanted to transmit. The majority argues that H. B. 1181 is yet more protective of publishers, because it turns the affirmative defense into an element—putting the burden on the State to show the absence of age verification measures. See ante, at 24–25. But in this context, the difference between an affirmative defense and an element is but a smidge: It will matter only when a jury thinks the presence (or absence) of age verification is a literal toss-up (which in the real world will be rare). And even if the difference is more than I think, it is one between two points on a continuum—not (as the majority insists) the dividing line between a “ban” and a “burden” on speech.

Much more important, the distinction between bans and burdens makes no difference to the level of scrutiny. When a statute draws lines based on the content of speech, strict scrutiny is required regardless of the amount of speech affected. Playboy made that point, in this context, at some length. “It is of no moment” to the level of scrutiny, the Court stated, that a law restricting speech “does not impose a complete prohibition.” 529 U. S., at 812 (emphasis added). And if that weren’t clear enough: “The Government’s content-based burdens must satisfy the same rigorous scrutiny as its content-based bans.” Ibid. And if that weren’t clear enough: When a statute regulates expressive content, no “special consideration” is given to the government “merely because the law can somehow be described as a burden rather than outright suppression.” Id., at 826. What’s more, Playboy is not alone in repudiating the majority’s reasoning. The refusal to countenance the ban/burden line the majority today peddles is fundamental to our free speech doctrine. Take any subject—say, because it is close to home, the Supreme Court. Ban speech about the Court; tax speech about the Court ($20 a pop); limit speech about the Court to certain times (Tuesdays and Thursdays); or (as here) demand identification to gain access to websites addressing the Court. Ban or burden, the level of scrutiny is the same: strict. See, e.g., Turner, 512 U. S., at 642 (stating the rule); Sorrell v. IMS Health Inc., 564 U. S. 552, 565–566 (2011) (same); see also, e.g., Reed, 576 U. S., at 159, 172 (adhering to the rule when reviewing mere burdens); Simon & Schuster, Inc. v. Members of N. Y. State Crime Victims Bd., 502 U. S. 105, 116 (1991) (same). So the principal distinction the majority draws between this case and the four that should control it is a non-distinction, by command of how we have always understood the First Amendment.

That leaves only the majority’s claim—again mistaken—that the internet has changed too much to follow our precedents’ lead. See ante, at 25–27. Of course technology has developed, both swiftly and surely. And that fact might matter (as indeed the burden/ban distinction might) to how strict scrutiny applies—and particularly to whether the State can show it has adopted the least speech-restrictive means to achieve its goal. Ashcroft explicitly recognized that point: It thought that, given the pace of technological change, the District Court might make a different decision than it had five years earlier about whether there were “less restrictive alternative[s]” to COPA. 542 U. S., at 671–672. To that extent—but to that extent only—the majority is right that Ashcroft was “self-consciously narrow and factbound.” Ante, at 26. Not, though, as to the level of scrutiny. On that question, the Court was unequivocal that because COPA was “a content-based speech restriction,” it must satisfy the strict-scrutiny test. 542 U. S., at 665; see supra, at 8–9, and n. 1. For that was a matter of basic First Amendment principle. And as this Court has understood: “Whatever the challenges of applying the Constitution to ever-advancing technology, the basic principles of the First Amendment do not vary.” Moody v. NetChoice, LLC, 603 U. S. 707, 733 (2024) (quoting Brown v. Entertainment Merchants Assn., 564 U. S. 786, 790 (2011)); see TikTok Inc. v. Garland, 604 U. S. ___, ___ (2025) (Gorsuch, J., concurring in judgment) (slip op., at 2) (“[E]ven as times and technologies change, ‘the principle of the right to free speech is always the same’” (quoting Abrams v. United States, 250 U. S. 616, 628 (1919) (Holmes, J., dissenting))).

Except that those basic principles do vary today.

III

The last part of the majority’s opinion—plus some of its footnotes—shows why all this matters. In concluding that H. B. 1181 passes constitutional muster, the majority states (correctly) that under intermediate scrutiny Texas need not show it has selected the least speech-restrictive way of accomplishing its goal. See ante, at 32. Even if there were a mechanism that (1) as well or better prevented minors’ access to the covered materials and (2) imposed a lesser burden on adults’ ability to view that expression, Texas could spurn that “superior” method. Ante, at 34. Likewise, the majority—because it is applying a more forgiving standard—can ignore a host of questions about how far H. B. 1181 burdens protected expression. See Tr. of Oral Arg. 67–68. In the fine print of two footnotes, the majority declares that it has no need to explore (1) whether H. B. 1181 requires covered websites to demand age verification for all their content or only for the subset that is obscene for minors; (2) whether H. B. 1181 requires that covered speech be obscene “only to a minor (including a toddler)” or “to all minors (including 17-year-olds)”; and (3) whether H. B. 1181 permits websites to use “newer biometric methods of age verification, like face scans,” that pose fewer privacy concerns than submitting government ID and transactional data. Ante, at 17, n. 7 (emphasis in original); ante, at 34, n. 14. The majority explains that even if Texas answered each of those questions in a maximally burdensome way—requiring government ID to view speech that is protected even for children because one-third of the website’s contents are obscene for two-year-olds—H. B. 1181 can go forward. And again, that is true even if Texas has a less burdensome way of “equally or more effective[ly]” achieving its objective. Ante, at 34.

I would demand Texas show more, to ensure it is not undervaluing the interest in free expression. Texas can of course take measures to prevent minors from viewing obscene-for-children speech. But if a scheme other than H. B. 1181 can just as well accomplish that objective and better protect adults’ First Amendment freedoms, then Texas should have to adopt it (or at least demonstrate some good reason not to). A State may not care much about safeguarding adults’ access to sexually explicit speech; a State may even prefer to curtail those materials for everyone. Many reasonable people, after all, view the speech at issue here as ugly and harmful for any audience. But the First Amendment protects those sexually explicit materials, for every adult. So a State cannot target that expression, as Texas has here, any more than is necessary to prevent it from reaching children. That is what we have held in cases indistinguishable from this one. And that is what foundational First Amendment principles demand. Because the majority departs from that right and settled law, I respectfully dissent.