Oral argument: Volokh v. James

1Next case, 23-356 is Volokh v. James May it please the Court, Sarah Coco for appellant, the Attorney General. General Business Law Section 394 of CCC imposes two straightforward requirements on for-profit social media networks doing business in New York.

2First, networks must provide a mechanism that consumers can use to communicate concerns to the network. And second, networks must disclose to consumers how they will respond to those concerns. Can I, I'm sorry, I'm jumping right into the beginning, which is rude, but I think the statute requires, drives so much of the analysis here.

3And when I read the statute, it didn't look like the sort of net choice you have to create a mechanism of disclosure, disclosure of content moderation policy. It seemed to be very specific to a specific kind of concern and a specific kind of report. Is it, is it your contention that we should read this statute as requiring only that users have an opportunity to provide feedback and that generic content moderation be disclosed, policies be disclosed without any indication in the framework that the feedback that consumers are being invited to give involves hateful conduct as defined in the statute?

4Yes, Your Honor. So I would distinguish there between the report mechanism requirement and the policy disclosure requirement. So our interpretation of the report mechanism requirement is that it does not require a mechanism that is exclusive to reports of hateful conduct.

5So as long as you can accept through the mechanism reports of hateful conduct, that's sufficient under the statute. Does it have to be identified as a report mechanism conduct for hateful conduct, even if it's not an exclusive portal? Does it have to be specifically identified as the place where those, that feedback goes?

6So our view of the statute is that for the report mechanism requirement, it does not. That's how we interpret the statute. And I would note that Professor Volokh has this blog post, which is in the Joint Appendix at 120.

7He agrees with us that that's the plain reading of the statute. As to the policy disclosure requirement, which is the second requirement, that does need to be disclosure of a policy about hateful conduct. It needs to, it needs to encompass what the network will do with reports of hateful conduct.

8So that policy can talk about anything the network wishes. Of course, they can take any action they wish or don't wish to take on hateful conduct. But a consumer would need to be able to look at that policy and say, okay, I understand what the network is going to do if I report hateful conduct as defined by the statute.

9So it'll have to reference, the policy disclosure will have to reference in some way hateful conduct as defined by the state? I don't think that's necessarily the case, and it's because networks could comply with us in a number of different ways. So to take Rumble's policy, for example, which the district court referenced, Rumble currently says that it will remove some types of speech, including but not limited to racism, anti-Semitism, and hatred.

10So that tells the consumer some things, but if the consumer wants to know what Rumble is going to do, for example, with content that would vilify someone based on their gender identity or gender expression, that policy doesn't answer the question. You don't think a reasonable observer would assume that they don't have a policy that covers that? Well, we think if Rumble expressly said that, that would be enough.

11And that's why we think there's really not a significant burden on the networks here. So if they said, we're not going to act on any other reports, then the consumer is going to know, I shouldn't expect a response to my reports about content that vilifies on the basis of gender identity, and we would think that's enough. So it's not required that they reference the definition of hateful conduct or even the term hateful conduct, but they do have to give the consumer that information.

12That's really the key. And so we would distinguish between the policy disclosure requirement and the report mechanism on that basis. So as to the report mechanism, we're arguing that plaintiffs lack standing to challenge this mechanism because they each already have available on their website an email address that can be used by consumers to report complaints, including incidents of hateful conduct, and none of them dispute that they have that available.

13And that's all the statute requires. And so because their current content- When you say including hateful conduct, do you mean that they reference hateful conduct? No, Your Honor.

14So our position- No. I'm sorry? Yeah.

15They don't necessarily reference what you call hateful conduct. That's correct. But our position is that the statute only requires them to have a mechanism that can be used by users to report those incidents, and they all allege in their complaints that they are already receiving these types of reports.

16In fact, one of their allegations is that, you know, if we're required to have this policy, we're going to receive even more of these. We already received some, and we don't want to receive them. And your view is nothing in the policy requires the network to respond to a report that they receive?

17That's correct. That's correct. And I think that's very clear in the text of the statute.

18In fact, the district court said that three separate times in the opinion. Can I ask about what I think is not talked about, which is sub 4? Yes, Your Honor.

19Okay. And it's the- So sub 4 is don't construe this to infringe on the First Amendment. Yep.

20It's as helpful as a blog post from Professor Volk. Sub B says to add or increase liability of a social media network for anything other than the failure to provide a mechanism to report and to receive a response on such a report. Yes.

21What does that mean? So our position is that the mechanism must allow the network to respond. So you can't have a mechanism where the consumer sends a report and it kind of just goes into the ether.

22The network, if it wanted to, would have to be able to respond. And of course, an email address satisfies that because you can always respond to an email. But they're not actually required to provide a response.

23I understood that to be- Maybe I'm- I understood that to be your argument about sub 2 and shall allow the social media network to provide a direct response. Allow. It, I think, captures the meaning that you just discussed.

24It's not required. Yes. Yes.

25But I'm wondering about how you get that out of sub 4, which says we don't increase liability unless you fail to receive a response on such a report to the consumer. It's anything other than the failure to provide a mechanism that would allow the user to report the hateful conduct and receive a response. We think it's exactly the same as sub 2.

26It's not trying to sneak in a separate response requirement. You know, that would be an odd place for the legislature to put a whole separate requirement. So you're saying provide a mechanism modifies, it's providing a mechanism to receive a response on such report?

27That's correct. So, so if I'm understanding it then, the statute says you don't have to respond to any of these, but you have to build in the technology that would allow you to do that. Is that what you're saying?

28That's correct. So if your policy is we will not respond, they can be in violation because the technology that they're using on their platform wouldn't allow them to if they wanted to. If they were using a mechanism that would not allow them to respond, for example, if the consumer reports just went into the ether and could not be responded to, that would be a violation.

29How does that make sense as an understanding? I mean, you know, just in terms of this notion of giving meaning to the words that the legislature use, like how does it make sense to say you have to have this technological capability, but we're completely agnostic as to whether you actually have to use it? Yes, Your Honor.

30So I think that's a question about what the legislature's rational basis for enacting this would be. And the legislature reasonably concluded that requiring networks to have this mechanism to report hateful conduct could result in a mitigation and violence in some circumstances. And here, of course, the legislature was reacting to the tragic event of the Buffalo shooting.

31And during that shooting, the shooter live streamed his crime on a social media network. And the live stream was interrupted when a user reported it. And so the legislature was looking at that and saying, if we allow users to report it, we're not going to require networks to respond.

32But networks may face pressure from consumers to respond, or they may choose to respond in certain instances. But allowing the user to report it could result in a mitigation and violence. And that is a sufficient rational basis for the legislature's decision to include this report mechanism requirement.

33So it sounds like this mechanism would facilitate liability sought by victims of hateful conduct by establishing that this website has the ability to receive these complaints and either did nothing pursuant to their policy or didn't do anything, even though they said they had a good policy, but basically to facilitate liability for failure to remove hateful speech. Well, Your Honor, nothing in the statute creates that liability, certainly. Well, I mean, let's just take a hypothetical example.

34I mean, somebody claims that they have been vilified because of their belonging, because they're Romanians, and then they are attacked, or the person who attacked them opens fire or commits other acts of violence. Couldn't the victims of that violence, as a result of this statute, sue the website and say, you had a mechanism for receiving these complaints, and you either did nothing pursuant to a policy to do nothing, or you had a policy to do something and you didn't do it? Well, Your Honor, I think if victims did do that, perhaps they could try to use the statute to allege a tort duty, but they would run into Section 230 immunity.

35And so I think that that would ultimately be addressed by a separate issue, but certainly nothing in the statute is intended to create a cause of action for that type of complaint. But it could... I'm sorry?

36It could create a risk that this liability could trump Section 230. Well, I don't think this... And cause people to do the state's bidding in suppressing what the state defines as hateful speech.

37Well, Your Honor, because Section 230 is a federal statute, I don't think we would dispute that that would preempt the state law to the extent... 230 is based on the idea that it's really impossible for people to sort through the on their websites. Here, however, you seem to have obviated that, for better or worse. But you seem to have obviated that, so it is no longer a matter of content moderation for your entire website.

38It's simply a matter of content moderation for those communications of which you receive notice. Well, Your Honor, perhaps that could arise in a different case. Nothing in the statute creates that cause of action.

39And if I could just briefly address the policy... I said facilitate. I didn't say cause of action.

40Yes, Your Honor. Perhaps. I do think there would be a Section 230 problem, but I suppose it would depend on the specific facts that were alleged.

41Yeah, let's shift. Minutes on the policy disclosure, which is maybe where the more robust action might be. Yes, Your Honor.

42So the policy disclosure requirement is subject to relaxed scrutiny under Zouder. And that's because the First Amendment does not prevent states from requiring truthful, accurate disclosures about the terms under which a good or service are going to be available. So here, the policy disclosure is requiring the networks to provide consumers with accurate information about the terms under which the service is going to be available, what the network.

43And that's consistent with many other policy disclosures in the law that require online entities, for example, to disclose things like their privacy policy. Obviously, the Fifth and Eleventh Circuit upheld very similar disclosures in the net choice cases. Do you think that our constitutional analysis of the policy disclosure is indistinguishable from the analysis of the disclosures in jurisdictions who have enacted laws that say, you must publish, you must disclose to the user your content moderation policies generically without reference to the specific subject matter of hateful conduct?

44So I can see why the fact that this is a policy disclosure on a specific subject could alter the analysis. I would just point out that that was also true in the Eleventh Circuit. And so the Florida law there said that the networks had to disclose their standards for censoring, deplatforming, and shadow banning.

45So I think that's exactly identical to what we have here. Those are actions, not content, right? Those are steps that the platform might take as opposed to the types of speech with respect to which the platform might take action.

46Well, I think it's similar because the network is disclosing a policy about what action it's going to take on hateful conduct. What they have to disclose is what they're going to do with the user reports. And so they don't have to disclose what their view is on hateful conduct.

47They just have to say, we're going to react to these reports or not. So in the state's view, does it matter how hateful conduct is defined in order to alter the constitutional analysis? Or can any definition of hateful conduct, so long as everything else is the same, doesn't change the analysis as to whether it's a policy disclosure on desire?

48Well, I think it would matter because it goes to whether the policy disclosure furthers an appropriate state interest. So as to whether it's, I guess, maybe a more specific question is to ask whether it's, not whether it meets the test, but whether, I guess what I want to say is whether it's speech or conduct. But I think you're answering it differently.

49If it's with respect to the disclosure, it's simply a matter of, does it meet intermediate scrutiny? And it could affect that, you would say. I'm not totally sure I followed the question.

50I'm sorry. Well, do you think the definition of hateful conduct, let me start here, impacts whether it regulates speech or conduct? I suppose that's a question for the reporting mechanism more than the disclosure.

51Yes. So as to the disclosure requirement, we're not arguing that that is a regulation of conduct. We agree that that's a regulation of speech.

52It's compelled commercial speech. And so that's our position on the disclosure requirement. And where the hateful conduct definition comes into the analysis is in whether it furthers an appropriate state interest.

53And so if the state had no interest in giving consumers this type of information, then it wouldn't further an appropriate state interest, and it wouldn't meet this outer analysis. But here, of course, we believe that this disclosure of the policy about hateful conduct does meet an appropriate state interest. It meets two.

54One is in transparency and informing consumers with accurate information. We know consumers are interested in this information, that they want to know this information. NetChoice has said in their amicus brief in this case that this is a subject that consumers are particularly interested in.

55You can also see that in the fact that many large social media networks already do disclose this type of information to consumers. So platforms have to provide truthful, accurate statements as to how they're going to treat this material. That's correct.

56What if what they say is not truthful? What if they say, we're not going to check this stuff? But they do, and they de-platform people.

57Or they say, we're going to make a conscientious effort to remove these things, and they don't. So I think then... What's the state going to do about it?

58So then they would be liable under the statute because... They would be liable to whom? For what?

59The Attorney General would be able to enforce the statute through her enforcement powers in the statute in Section 5, the civil fines and so forth, as to the network. Because they would not be disclosing accurate information to the consumer about what they do with these reports of hateful conduct. If their policy was a lie, then that would be...

60That would lead to liability under the statute. Well, if you lie about the ingredients of a cookie, that's one thing. But isn't there a lot of play in the joints as to what hateful conduct is?

61What is vilification? What is humiliation? What is actionable?

62And therefore, how can somebody make a statement that the Attorney General can deem true or not, accurate or not? Well, the statute does not require the networks to identify content that meets the definition of hateful conduct. All they have to do is say what they're going to do with those user reports.

63And so they don't have to identify any content as hateful conduct or not. And this is... They have to say what they're going to do about what?

64About reports of hateful conduct, which the state has defined in a definition that really doesn't seem to mean all that particular. So they would have to... If they're going to make an accurate statement, they're going to have to make an accurate statement about what they're going to do about receiving a certain kind of complaint.

65And that kind of complaint is not very clearly defined. I mean, you really don't think that that definition is definitive, do you? Well, Your Honor, all the networks have to do is be transparent about what they'll do with these user reports.

66So they can have any policy they wish their policy can be. But it has to be about hateful conduct. It has to address what they would do with the reports of hateful conduct.

67But as I said, Rumpel... So the definition of hateful conduct that the state has provided is built into that obligation to say what you're going to do about it and if that statement is not accurate or truthful, well, let's say not accurate, because the platform has a different view of what hateful conduct is, then they're liable for $1,000 a day penalty. So they would have to accurately disclose, that's true, what they are going to do with these reports as defined by the state definition.

68And so if that was not truthful, if they said we're going to act on these reports and then actually they really didn't, that would lead to liability under the statute. I think this has to do with the district court's vagueness analysis. And I want to point the court to one of the Supreme Court decisions we cite in our brief, which is the Rowan decision.

69And there the Supreme Court was looking at a law that allowed recipients of mail, if they deem the mail sexually provocative, to say to the U.S. Post Office, I don't want to receive this mail anymore, and the sender would not be able to send it to them. And the court said...

70And the sender challenged that saying it's vague. I don't know what the recipient is going to deem sexually provocative. And the court said it's not vague.

71Your obligations under the statute are clear because what you have to do is not send mail once the sender has identified it as sexually provocative. What the networks have to do here is say what they're going to do with the consumer reports. But the consumer is the one who's identifying whether they think the content is hateful conduct or not.

72And then the network can do whatever they wish with that report. They just have to tell the consumer what they're going to do with the report. Can I just circle back to where I left off with you about whether it matters what the definition is.

73And I think for policy disclosure, you have to also address the editorial discretion line of argument. That is to say whether it's forcing... And along the lines of Judge Jacobs, is it forcing an acceptance at some level of the meaning of hateful conduct?

74So let me change. Another state is concerned about children being exposed to positive images of LGBT people. So they define hateful conduct as anything that celebrates the lives of LGBT people.

75Everything else is exactly the same. How do you run the analysis? So I think under any standard of scrutiny, it would be very difficult for a state to show what the appropriate state interest was in not allowing children to see those images.

76Tell me how you run the analysis as to what the level of scrutiny is. So I think that would still potentially fall into the Zouder analysis. But even if it did, the state would not be able to show an appropriate interest.

77So there, I don't think the standard of scrutiny is going to be decisive because... And that's similar to, for example, if there was a policy requiring networks... Or excuse me, a regulation requiring networks to have a policy about, I don't know, anti-democratic speech or anti-Republican speech.

78Like the state always has to show an appropriate interest. And so I don't think the court should be worried that applying Zouder to these types of disclosures is going to lead to any disclosure being accepted. The state has to show an appropriate interest.

79And here, the statute was specifically enacted in response to a situation where these user reports did matter. That was a feature of the violence that occurred. And so this is not the state that's kind of enacting a policy out of the blue.

80Well, just one quick question. Do you have any view as to whether we should or should not wait for the court to act in that choice, they didn't take the disclosure questions, but presumably there could be discussion of the boundaries of content moderation? Yes.

81So we do think it's important that the Fifth and the Eleventh Circuit, although they disagreed about every aspect of the First Amendment analysis, agreed on this. And then the Supreme Court declined to grant cert on this. And so we think that's pretty strong evidence that our analysis is the correct one here.

82That said, obviously, the state recognizes that the Supreme Court may say things relevant to the court's decision in the net choice cases. And so while we do certainly have an interest in getting this decided quickly, there's a state law that's been preliminarily enjoined on its face. I think we would defer to the court on whether it'd be appropriate to wait.

83Do you think there could be some relevant discussion that could impact this? I can certainly imagine things the court might say. Yes, Your Honor.

84Thank you. Thank you. Mr.

85Diaz. Good morning, Your Honors. May it please the Court.

86James Diaz for the plaintiff. Appellees, the district court correctly enjoined New York's online hate speech law because the First Amendment does not just protect our right to speak, but also protects our right to say nothing at all. And here, what the New York statute says is that the New York statute demands that countless websites across the internet explain the contours of hate speech.

87It also requires that they speak about what types of moderation policy they're going to have towards this particular type of speech that is defined in the statute. The court should affirm the preliminary injunction, among other reasons, because it compels speech in a viewpoint discriminatory manner, it coerces the elimination or reduction of protected speech, it chills protected speech, and is unconstitutionally vague. Now, I understand the state to be arguing, oh, no, it doesn't actually require any reference to hateful conduct.

88It just requires a reporting mechanism and a disclosure of our policies that is, I guess, broad enough to encompass the conduct that might fit within that definition. If that were true, if it were basically a 5th, 11th Circuit style, a disclosure content moderation policies, would that be constitutionally problematic from your perspective? Well, of course, here, I'm not sure I fully understand the question, because the disclosure laws in Texas and in Florida are significantly different.

89Right, I understand. But there are ways in which the state's construction of the statute starts moving in the direction of those statutes. And so I just, in terms of grounding your argument, I want to understand whether you think that a requirement that a website post a generic content moderation policy, whatever their content moderation policy is, including we don't moderate content, whether that is constitutionally problematic.

90I'm trying to figure out whether the definition of hateful conduct and the incorporation of that definition into the statute is essential to your constitutional argument. I think, well, I think it is very important to our argument. But here, if we look to what this court said in Evergreen, for example, there you had pregnancy service centers who had to say whether or not they provided abortion referrals to people who came to them.

91Now, again, that was a neutral requirement whether or not they did something. And it was, but what the court said in that case was clearly they have to talk about abortion, number one, a controversial issue, and they have to, which is political in nature, and that they have to address this when there are at least less restrictive opportunities to regulate and achieve the state's interest. However, the court looked at it through the lens of strict scrutiny and even said under intermediate scrutiny such a policy could not pass because it will put in the mind of the pregnancy service center that which was not already there.

92That's what the Supreme Court has said cannot be done. So the First Amendment protects a right not just to speak, but to say nothing at all on particular political issues. And here we have hate speech.

93We're talking about something that has been debated, what it is, what the exact policy is, whether it's protected, what it means, that has been debated for over a century. And right now, over just the last month or two, we've been discussing around the country whether hate speech is actually, should be moderated online. This policy is asking for every website, as the state just said, to say whether or not it should be moderated on their site.

94But the state is saying that somebody wishes to be free of such a burden need only say our policy is to do nothing or our policy is none of your business. You know, your message is very important to us, but we're not going to respond to it. The issue there is that that is taking a position and it's forcing the websites to take a position on whether certain types of speech here, hate speech, that vilifies or humiliates certain protected classes, should be moderated or not.

95And it's saying, we don't believe it is, or here's how we think it should be done. That is inherently taking a position. And that's what is required by the statute, when what the First Amendment ensures, if we go to Riley, if we look to Pacific Gas, if we look at 303, that which is not in someone's mind, they do not have to say.

96And that's what the state is requiring of countless websites here with the mechanism and the policy. What about a policy that says we will respond to your complaints? I think this is similar to the policy that's published in our sole discretion.

97However, we choose to respond. That's our content moderation policy. Is that burdensome on the platform to have to disclose that?

98It's burdensome because if they have to disclose what they're, well, first, I want to talk about Volokh's post because he did not actually state what his policy is in that post. He talked about, he was pondering what would satisfy this statute. Perhaps it's this, but perhaps he would have six months to decide whether to challenge a lawsuit.

99And this lawsuit is his answer to that question. Now, what's clear is that there is a burden on websites for posting any policy related to hateful content. They may alienate their community, especially the plaintiffs here.

100They push their free speech message. They are here to preserve the marketplace of ideas. And if they have to talk about what hateful content is and how they deal with it, it may be viewed as, even if they say it's within their discretion, that means it's going to be looked at.

101It's going to be decided upon. It will not preserve the marketplace of ideas that they would like to see happen. Does it matter, for your view, sort of flip side of the question of the state, does it matter to what you just articulated how hateful conduct is defined?

102Well, hateful conduct here is defined. No, I understand. But does it matter, for your view, how it's defined?

103Could it be defined in a way that, with everything else the same, doesn't implicate any First Amendment concerns? Well, I think here what's clear is that hateful conduct has to be adopted in some form. No, I get that.

104So we could imagine changing language in hateful conduct. I mean, I suppose if it said just, if you cut out and just said, incite violence or something that captured incitement, do you have a First Amendment argument? I think, so if the question is whether, if the definition just encompassed, say, unprotected speech.

105Yeah. I think this Court can rule on our issue without going that far. Because clearly, we have an over-breath claim.

106Clearly, the definition here involves a substantial amount of protected speech. So, and I think the issue of whether this Court could address the request for an unlegal speech is unsettled. I'll give you a, so I'll try a specific hypothetical that's different than the question I asked.

107But let's say a state is, in the wake of October 7th, passes exactly the same, but it defines hateful conduct as advocating the genocide of Jews. Everything else is the same. How do you run the analysis?

108Well, I think it would, well, first you'd look at the context. Then you would say, is it speech? Is it not speech?

109What type of speech is it? So, go ahead. Commercial?

110Yeah. And I think here, it would have to be viewed as a viewpoint-based discrimination because it addresses a specific type of speech with a specific perspective and requiring websites to talk about that particular content. So, a state can't require social networks to have a reporting mechanism for the advocacy of the genocide of Jews and a disclosure policy as to how they'll deal with that.

111That would be unconstitutional. That is a particular viewpoint that people may hold or may not. It's not for the state to determine what should be in the marketplace of ideas outside of unprotected speech.

112So, if it would fall into one of those categories, perhaps. But here, in that case, we have a particular perspective, as hateful as it may be, as harmful as it may be, as hurtful as it may be, that is still nonetheless protected by the First Amendment and this Court's precedents. Now, what's clear is that it's not just that this is viewpoint discrimination and content-based discrimination on the face of the statute, but that it does, in fact, require the incorporation of the definition of hateful conduct into every website's policies.

113Whether it's stated on its face or not, it has to be there because the risk of prosecution here is serious. And I think you could say, I think you agree with this interpretation of the statute. Here's the state's definition of hateful conduct.

114We fundamentally disagree with that and we won't. Either here's our definition or we don't have a definition or we respond at our discretion. I mean, it could be speech that completely disavows the state's definition and then the network isn't forced to do anything.

115Do you accept that as a construction of the statute? It may be a construction of the statute, but it's not permissible under the Constitution. Because if that were the case, if all that needed to happen here was that a website could say, we disagree with this, but it's required by the state, then no compelled speech regulation would be unconstitutional.

116They would all be permissible. But having to add speech is a further problem here because it furthers the compelled speech problem and furthers the alteration of the content of the website. In Riley and Evergreen, as this Court agreed with in Evergreen, the alteration of content on a website necessarily is compelled speech and it necessarily changes the content of the website.

117In doing so, that is in itself the restriction on the right to freely communicate about particular ideas. I suppose we have requirements for websites now to have their privacy policies. And that's mandating words on the site that describe their policies to use of your personal identifying information, et cetera.

118Is that constitutionally problematic? It would depend on the context. So here we're talking about the speech of users and creators on the site.

119In Yuji Balog's case, that could include him as well. If we're talking about the privacy of users and whether they have to publish privacy policies because they do something with user data, that's a different question. I think that may involve commercial speech.

120It may involve certain commercial transactions. Here, we don't have that at all. This Court does not need to go that far in ruling in our favor because of the viewpoint-based nature of the statute.

121It is against a particular form of speech. The New York restaurant's case where they were required with calorie information. A, I was trying to figure out whether that's still good law after this era.

122But assuming it is, you have these commercial enterprises and it's, quote, purely factual speech, but it's speech they don't want to engage in. It forces them to engage with subject matter that they would prefer not to. And we upheld that.

123Why isn't this essentially that? Here, we do not have what Zotero requires, which is a purely factual and uncontroversial disclosure. There's nothing purely factual about a website stating its opinion on what is or is not hateful, what should be moderated.

124It just has to say, here's our policy. That's purely factual. It doesn't have to say what they think is hateful or not.

125What their policy is with regard to hateful speech as the state defines it may be a fact, but it is an opinion. It's not like mercury in light bulbs. It's not a scientific calculation like in calories in a cheeseburger.

126It's something that we all agree to. No, but the parallel is what the network's policy is, factually. Isn't that what's required?

127I don't think so. I think the actual information that is being disclosed has to be something that is purely factual and uncontroversial. And here, as this Court said in Evergreen with regard to abortion, what hate speech is, what it's defined as, what type of moderation policy should apply to it is anything but controversial, especially in the current era.

128So I understand the controversial point, but I still don't understand. I'd like to, the argument that it's not simply requiring a factual statement of what the network's policy is. Well, here, what the law asks for is not the policy, any policy related to what it thinks is hateful.

129It's asking for what its policy is and what they will do with speech that the state thinks is hateful. That is forcing websites to adopt the message of the state and that what the state thinks is hateful is not a fact. It is an opinion and it is a vague one at that.

130If we look at what the state is saying in its definition, what vilifies or humiliates, it's very difficult to know what actually is being regulated on the website. So the plaintiffs, if they had to adopt this definition, would have no idea whether it actually meets the standard. They don't have to adopt the definition again.

131I think that you said they wouldn't know what to do if they have to adopt it, but what statutorily requires them to adopt it? I think two things. First, when we have the policy and the mechanism, the policy itself has to be in reference to hateful conduct, which the state defines, we call hate speech.

132I think the state mislabels it. And the policy also refers to what the mechanism is for. Again, hate speech.

133So, and as the state said, what they require is that what is being used and what is being discussed when a website talks about what is hateful has to comport with the state's definition. Do you have a view as to whether we should wait for net choice? I think this court certainly could.

134I think there will be portions of that decision that will be relevant. However, I don't think it's necessary here. While the intrusion into editorial discretion is one potential reason that this court should not overturn the preliminary injunction, it's not the only reason.

135And here, what we have goes much further than just an intrusion into the editorial process and decision making and requiring certain content be posted on a website. We have what clearly is a requirement not just to state something generally or something that is already in existence, but to state something new. To state we now have a policy about this specific type of speech.

136Here it is, and here's where you can report it. Can you tell me, there's this discussion about the as-applied challenge versus a facial challenge. And usually when you have like, oh, this is an as-applied versus facial, there's something about the particular litigants that might distinguish them from the universe as a whole.

137Is there any universe in which this statute is unconstitutional as applied to your clients, but there may be some other application in which it would be constitutionally applied? I don't think so. And even if there was, the overbreath claim here and the vagueness claim make that a moot point in the end.

138Because what we have is not just an impact on our client's speech, but also the speech of all of their users and among users in other places. And which also leads to the facial, which also is a facial claim against the statute. Thank you.

139Thank you. So I'd like to start with what zowder means by factual and uncontroversial. And those two terms should be read in conjunction to mean a disclosure that is truthful, that is accurate, that is not misleading to the consumer, and that is not a disclosure of opinion.

140But simply because something touches on a matter of public debate, which is exactly what my friend just said, does not mean that it is not commercial speech subject to zowder. And this court has said that in Connecticut Bar Association. And that's taking language directly from zowder.

141So the fact that something is a subject of public debate does not take it out of the zowder context. In addition, I want to briefly address Evergreen. So the court there addressed zowder only in a footnote.

142It said in dicta that abortion is a controversial service, but it did not explain what it thought factual and controversial meant pursuant to zowder. And so I don't think the court should rely on that. But even if it did, the crisis pregnancy centers in Evergreen and in NIFLA are not similarly situated to the networks here.

143Because the networks here are already discussing this subject. Rumble has a policy that says that we will remove content that is grossly offensive to the online community because it's hatred. And so this is not like a crisis pregnancy center that doesn't disclose anything about its policy.

144Does that disclosure satisfy the statute? It does not, Your Honor, only because they don't say what they would do with other reports. So if Rumble added we will not act on any reports other than the reports listed here, that would satisfy the statute.

145But their current policy doesn't because you'd have some reports under the statute where the consumer wouldn't know what the network was going to do with them. Suppose, I mean, the mechanism part of the statute, you say, doesn't really distinguish. People can make complaints about anything.

146They can complain about things that vilify Romanians. They could complain about pornography. They could complain about cruelty to animals.

147And they complain about hateful conduct. Those, there's just all of those different inputs. Does the policy have to address hateful conduct in particular?

148And is it required only to address that? I mean, do they have to also disclose what they'll do with all of these other kinds of complaints? So the networks can disclose any information they want in addition to the subject of the policy.

149But what they must address is enough, they must provide enough information that the consumer is going to know how they're going to act on the consumer's report of hateful conduct. So they have to say what we're going to do with those reports. And then if they want to say we're going to act or not act on all these other topics of speech, that's absolutely fine.

150And that's a hallmark of a souder. If they have a policy, if they have a policy on animal cruelty, does the law require them to disclose it? Certainly not.

151Certainly not. Nothing in the statute addresses that. I want to briefly address the allegation that the statute is viewpoint-based.

152It is not viewpoint-based because the regulatory burden of the statute does not turn on the regulated entity's viewpoint. So here, if a network has a policy of removing hateful conduct, the burden is exactly the same as if the network has a policy of never removing hateful conduct. And that's different than something like Mattel versus Tom, where the federal government was saying whether or not you get the trademark protection turns on your speech, turns on whether the mark is disparaging or not.

153So here, nothing turns on the viewpoint of the network. And that's why it's not viewpoint-discriminatory. On the editorial discretion question, we don't think that comes into this analysis because nothing in the statute infringes on networks' editorial discretion because it doesn't require them to remove any user content.

154That's also why the overbreath analysis that the district court did is flawed. Nothing here is telling networks that they have to remove user content. Nothing is preventing users from speaking in any way.

155The statute does not regulate users or user content. It doesn't require networks to do anything with that content. And then briefly, two additional points.

156First of all, if the court concludes that there are ambiguities... If I may, I'm sorry. But if the policy says that we are going to de-platform people who engage in hateful conduct, isn't that going to have an impact on the people who might wish to post their views on that website?

157It might have an impact, but that impact is because of the network's policy, not because of the statute's requirement that the network tell the user what that policy is. But the network didn't say what the policy was, and people might not be chilled in doing things. Not that people should go around doing hateful things, but I don't exactly know what hateful is.

158And people may say, under that definition, I don't know whether what my thoughts are would be characterized by the legislature of New York as being hateful. So I'm not going to express those thoughts. So if the network has that policy, even if they don't disclose it to the user, maybe the consumer will be surprised when they're de-platformed, but they're still going to be de-platformed.

159And that's what's going to chill their speech. It's not the disclosure requirement. It's the fact that the network is going to de-platform them.

160And that's separate from any regulation imposed by the state. And that's why users would not have standing to challenge this law and why the district court's over-breath analysis is flawed. Over-breath is an exception to the third-party standing rules.

161And so you need to bring in the interests of a party not before the court that would have standing to challenge the statute. And here, users would not have standing to challenge the statute. And we cite the Meinzing versus Bonta district court case, which analyzed this issue as to a similar California law in our brief.

162So just very briefly, two final things. First of all, if the court thinks that the analysis here turns on an ambiguity in the statute, and we do think that the statute is clear, but any ambiguity would suggest that this question should be certified to the New York Court of Appeals. And that's because no New York court has ever had the opportunity to address this statute.

163It's a duly enacted New York statute. I thought your argument was constitutional avoidance in your papers, if there's ambiguity and certification. Yes.

164So constitutional avoidance would certainly apply, and also certification. We have both in our papers. And so we don't think that an ambiguity should result in affirming the injunction.

165And then second, I just want to emphasize separability again. And so all of my friend's arguments were focused on the policy disclosure requirement. That's the only thing the district court analyzed.

166And here there are two separate requirements that should be analyzed separately. Thank you. Thank you.

167Thank you. Appreciate it. Well argued, both of you.

168And that concludes today's cases. And we'll ask the court to close.