Plaintiffs, rightsholders of musical recordings, all affiliates of EMI, appeal from the judgment of the United States District Court for the Southern District of New York (Ronnie Abrams, J.) granting summary judgment to defendants, Vimeo, Inc. and Connected Ventures, LLC (collectively “Vimeo”), dismissing Plaintiffs’ claims of copyright infringement on the ground that Vimeo is entitled to the safe harbor provided by Section 512(c) of the Digital Millennium Copyright Act (“DMCA”), 17 U.S.C. §512(c), which, in certain circumstances, protects internet service providers from liability for infringement when users of the service upload infringing material onto the providers’ websites. The district court rejected Plaintiffs’ contention that Vimeo had actual or red flag knowledge of the infringement or the right and ability to control the infringing activity, and therefore lost entitlement to the safe harbor. AFFIRMED. PIERRE LEVAL, C.J. This is an appeal by Plaintiffs, who are rightsholders of musical recordings, all current or former affiliates of EMI, from the judgment of the United States District Court for the Southern District of New York (Ronnie Abrams, J.), granting summary judgment to defendants, Vimeo, Inc. and Connected Ventures, LLC (collectively “Vimeo”). The district court dismissed Plaintiffs’ claims of copyright infringement. Section 512(c) of the Digital Millennium Copyright Act (“DMCA”) establishes a safe harbor, which protects qualifying service providers from liability for infringement when users of the service upload infringing material onto the providers’ websites. See 17 U.S.C. §512(c)(1). However, the safe harbor is not available to a service provider if the service provider (A) has actual or red flag knowledge that the material on its website is infringing and fails to remove the infringing matter expeditiously, or (B) has the right and ability to control infringing material on its website and receives a financial benefit directly attributable to that activity. See id.1 Vimeo is a qualifying service provider and operates a website on which users can upload videos to share either within a restricted circle of users or with the general public. At issue in this appeal are 281 videos containing copyrighted musical recordings owned by Plaintiffs. All of the videos selected by Plaintiffs to place in issue in this suit are videos with which Vimeo employees interacted after those videos were uploaded by users to Vimeo’s website — for example, by selecting the video to be featured in a prominent section of the website or by posting a comment about the video. The district court found that Vimeo was entitled to the DMCA’s safe harbor because: (1) although there was evidence that Vimeo employees had interacted with videos containing infringing content, there was insufficient evidence to prove that it would have been obvious to those employees that the content of the videos was neither authorized by the rightsholder nor fair use; and (2) Plaintiffs failed to show that Vimeo had sufficient “right and ability to control” within the meaning of the statute to lose entitlement to the safe harbor. We agree with the district court that Vimeo is entitled to the safe harbor and therefore AFFIRM the judgment. BACKGROUND 1. Facts A. The Digital Millennium Copyright Act (“DMCA”) “The DMCA was enacted in 1998 to implement the World Intellectual Property Organization Copyright Treaty and to update domestic copyright law for the digital age.” Viacom Int’l, Inc. v. YouTube, Inc., 676 F.3d 19, 26 (2d Cir. 2012) (internal quotations and citations omitted). Title II of the DMCA, the “Online Copyright Infringement Liability Limitation Act” (“OCILLA”), established four safe harbors that allow qualifying service providers to limit liability for certain claims of copyright infringement. See id. at 27. We have described these safe harbors as a “compromise” between protecting copyright owners and “insulat[ing] service providers from liability for infringements of which they are unaware, contained in material posted to their sites by users, so as to make it commercially feasible for them to provide valuable Internet services to the public.” Capitol Recs., LLC v. Vimeo, LLC, 826 F.3d 78, 82 (2d Cir. 2016) (“Vimeo I”). If a service provider meets the threshold qualification criteria, see 17 U.S.C. §512(k)(1)(B), (i)(1)-(2),2 it then has passed those tests of eligibility for one of the four safe harbors. The safe harbor at issue in this case is provided by Section 512(c), which protects the provider from liability for infringement that would arise “by reason of the storage at the direction of a user of material that resides on a system or network controlled or operated by or for the service provider.” Id. §512(c)(1). The statute provides (as here pertinent): [(c)(1)] A service provider shall not be liable for monetary relief…for infringement of copyright by reason of the storage at the direction of a user of material that resides on [its website]…, if the service provider — (A) (i) does not have actual knowledge that the material or an activity using the material on the system or network is infringing; (ii) in the absence of such actual knowledge, is not aware of facts or circumstances from which infringing activity is apparent; or (iii) upon obtaining such knowledge or awareness, acts expeditiously to remove, or disable access to, the material; (B) does not receive a financial benefit directly attributable to the infringing activity, in a case in which the service provider has the right and ability to control such activity; and (C) upon notification of claimed infringement…, responds expeditiously to remove, or disable access to, the material that is claimed to be infringing or to be the subject of infringing activity. Id. §512(c)(1)(A)-(C).3 The statutory scheme also has a notice-and-takedown provision: Section 512(c)(2) requires that service providers designate an agent to receive notifications of claimed infringement, and Section 512(c)(3) explains what elements must be included in such a notification. See id. §512(c)(2)-(3). Upon receipt of a valid notification, Section 512(c)(1)(C) requires that the service provider “expeditiously” remove or disable access to the infringing material. See id. §512(c)(1)(C). B. The Vimeo Website Vimeo is a video-sharing platform created in 2004. It has become one of the most popular websites on the internet. In 2007, Vimeo had only 40,000 registered users, but as of 2012, it had about 12.3 million registered users posting about 43,000 new videos a day. Users who have registered an account with Vimeo can upload videos to the site. They can add tags (i.e., keywords associated with the video) or descriptions (also called captions). Other users may write comments, signal that they “like” videos, subscribe to interest groups that will share videos of preferred types, and create or subscribe to channels, which are repositories of videos that have been grouped together by a common theme or category. Registering as a user is free and can be done anonymously.4 A posting user may choose to designate a video as “private,” in which case the video can be viewed only by accounts designated by the poster or those who possess a password selected by the poster. If a video has not been designated as private, both registered and non-registered users (essentially anybody with access to an Internet connection) can play the video on demand — either directly on Vimeo’s website or through a third-party website where the video is embedded — or download the video for later viewing.5 Vimeo offers both free and paid subscriptions. The majority of its revenue comes from user subscription fees. Vimeo also earns revenue from the placement of advertisements on its website, in two ways: for some advertisements, Vimeo contracts with the advertiser and typically is paid each time the advertisement is viewed; for others, Google’s AdSense program selects which advertisements to display to which customer, and Vimeo is paid each time a user clicks on the advertisement to go to the advertiser’s website. Beginning in 2011, by contract with certain rightsholders in copyrighted recordings (mostly smaller labels, not including any Plaintiff in this suit), Vimeo began offering licenses to users to reproduce copyrighted music through its music store. Vimeo earns commissions through such sales. C. Background of the Dispute Some videos uploaded by users onto Vimeo’s website incorporate recordings of musical performances for which the rights are owned by Plaintiffs (including the “Videos-in-Suit”). This lawsuit seeks to impose liability on Vimeo for those alleged infringements. Vimeo, for purposes of these summary judgment motions, conceded that the Videos-in-Suit contain music owned by Plaintiffs and that the uses in those videos were infringing uses. See J. App’x at 312-13 & n.3. It is uncontested that Plaintiffs have never licensed this music to Vimeo or its users who posted it and that Vimeo has never paid Plaintiffs for the use of the music. In 2008, EMI sent Vimeo a cease-and-desist letter demanding removal from the website of videos containing EMI recordings and compositions. The letter enclosed a list of 173 videos to be removed from the site and demanded that “Vimeo immediately take appropriate action to ensure that all other EMIowned or-controlled works are removed from the Site.” J. App’x at 104. Vimeo took down the 173 videos listed in the letter. Following this incident, Vimeo employees emailed one another, seemingly mocking EMI, referring to EMI personnel as “dicks” and “goofballs.” J. App’x at 994. This lawsuit was filed a year later. Since then, EMI has sent additional take-down notices, with which Vimeo has complied. On appeal, there are 281 videos in contention. These are videos uploaded between 2006 and 2013, which include recordings of musical performances owned or controlled by Plaintiffs, and with which Vimeo employees interacted (for example: commented upon, liked, promoted or demoted, or placed in a community channel). These do not include any videos that were uploaded by Vimeo employees. D. Vimeo’s Upload Requirements, Curation, and Moderation Unlike other video-sharing websites, Vimeo does not allow users to upload videos that were not created — in whole, or in part — by the uploader. Vimeo tells users that it does not allow the posting of specified types of videos, which include advertising videos (such as commercials, infomercials, and product promotions), videos containing abusive (e.g., bullying) or sexually-explicit content, real-estate walkthroughs, and, since 2008, so-called “gameplay” videos, wherein the video exhibits a person playing a video game.6 Vimeo’s rules also disallow uploads of movies, TV shows, and trailers. In order to register an account, users must agree to abide by Vimeo’s Terms of Service. Those terms include an agreement not to upload videos that infringe another’s rights. Every time a user uploads a video, the user is reminded of the agreement to upload only self-created videos and to not upload videos that are intended for commercial use or are otherwise inconsistent with Vimeo’s content restrictions. In practice, however, users can generally upload any videos without interference from Vimeo. Vimeo does not pre-screen videos presented for uploading. It does engage in a small degree of moderation and curation of its website. As examples, members of Vimeo’s “Community Team” occasionally post “likes” (a symbol, such as a heart, indicating approval) or comments, upload their own videos, and create or subscribe to groups and channels. They also promote user videos by putting them in prominent places on the website, such as on Vimeo’s own blog, or on the “Staff Picks” channel. In some instances, staff members remove videos (and, at times, entire user accounts) for violation of the Terms of Service. Vimeo has also encouraged users to create certain types of content. In 2006, Vimeo’s founder created a video of himself lip-synching to music while walking down the street, and then synced the video with music. He named it “lip-dubbing” and invited others to create their own lip-dubbing videos. At a company party, Vimeo’s employees created their own lip-dub video, which was watched millions of times. Lip-dub became a popular trend, and Vimeo created a channel under the name “Lip Dub Stars.” A lip balm company paid Vimeo for the opportunity to advertise on the lip-dub channel and also funded a lip-dub contest. Vimeo employs various computer programs (“Mod Tools”) to assist the Community Team in locating and removing videos whose content violates the Terms of Service. One tool identifies videos that have recently received a large number of views; another searches the website for current film titles; another identifies videos that are approximately the length of a typical half-hour or hour-long television show; and another identifies users who are uploading a large number of videos in a short period of time. When videos and/or users are identified by one of these tools, Vimeo staff manually review them. Vimeo also enables users to “flag” videos that they believe violate the Terms of Service. Community moderators evaluate the flagged content and decide whether or not to remove it. E. Employees’ Knowledge Concerning Licensing and Copyright Laws Plaintiffs have offered evidence to suggest that Vimeo employees had particular experience with and awareness of facts that would allow them to discern whether posted uses of music were licensed or were fair use. Some Vimeo employees have published their own videos, during the time period of interest, including videos containing Plaintiffs’ music, without securing a license. In or around 2009, Vimeo employees were told by Vimeo’s legal team not to use copyrighted music in the background of videos they created. In answering questions from users about whether copyrighted music could be included in the background of a video, Vimeo’s pre-scripted response was that “adding a third party’s copyrighted content to a video generally (but not always) constitutes copyright infringement under applicable laws.” J. App’x at 75, 175, 404-05. Plaintiffs provided evidence showing that, on at least a few occasions, Vimeo employees added text that undermined the official message, such as, “off the record…Go ahead and post it. I don’t think you’ll have anything to worry about.” J. App’x at 175, 404-05. Vimeo characterizes these interactions as “few in number, unauthorized, and…not reflect[ing]…Vimeo’s policy.” J. App’x at 448-49. In 2011, Vimeo launched a virtual music store where users could pay a fee to obtain a license to use certain non-major-label music. In launching the store, Vimeo published a blog post which explained that Vimeo employees know that licensing music can be “confusing” and “painful.” J. App’x at 1130. 2. Procedural History On December 10, 2009, Plaintiffs filed their initial complaint, which named 199 videos that contained allegedly infringing uses of music licensed to Plaintiffs. In May 2012, Plaintiffs moved to amend the complaint by adding over a thousand videos. The district court denied the motion with leave to refile after the court ruled on a motion for summary judgment. On September 7, 2012, Vimeo moved for summary judgment asserting entitlement to safe harbor protection under the DMCA, and on November 16, 2012, Plaintiffs cross-moved for partial summary judgment that Vimeo was not entitled to the safe harbor. On September 18, 2013, the district court ruled that Vimeo did not have the right and ability to control the infringing material at issue and therefore was not excluded from the safe harbor on that ground. However, it also ruled that a triable issue of material fact existed as to whether Vimeo had knowledge or awareness of infringing content for 55 of the videos, with which Vimeo employees had interacted.7 As to the remaining 144 videos, the court granted partial summary judgment to Plaintiffs on those videos containing infringed-upon material recorded prior to 1972, because, in the district court’s view, the DMCA safe harbor did not extend to recordings made prior to 1972. It granted summary judgment to Vimeo on the remaining videos. On reconsideration, the district court granted summary judgment to Vimeo on an additional 17 videos. The court also granted Plaintiffs leave to amend the complaint, adding 1,476 videos, and certified three questions for interlocutory appeal: (1) whether the DMCA safe harbor extended to recordings made prior to 1972; (2) whether Vimeo had knowledge or awareness of infringing content; and (3) whether Vimeo had “a general policy of willful blindness to infringement of sound recordings.” Vimeo I, 826 F.3d at 82. On that interlocutory appeal, this court held that the DMCA safe harbor does apply to pre-1972 recordings, and thus vacated the grant of partial summary judgment to Plaintiffs on the videos alleged to infringe music from before 1972. See id. at 93, 99. We also vacated the district court’s denial of summary judgment to Vimeo on Vimeo’s red flag knowledge of infringement, ruling that Vimeo was entitled to summary judgment “unless plaintiffs can point to evidence sufficient to…prov[e] that Vimeo personnel either knew the video was infringing or knew facts making that conclusion obvious to an ordinary person who had no specialized knowledge of music or the laws of copyright.” Id. at 98. We affirmed the district court’s finding that Vimeo had not been willfully blind to infringement. See id. at 98-99. On remand, the parties identified 307 videos remaining in dispute and renewed their cross-motions for summary judgment. The district court ruled in favor of Vimeo, granting summary judgment in its favor on 281 videos, holding that, under our standard laid out in Vimeo I, Plaintiffs had failed to show that Vimeo had red flag knowledge of infringing content. On the other hand, the court did find that there was a disputed issue of material fact as to 26 of the allegedly infringing videos, which had been uploaded by Vimeo employees. To permit the district court to enter an appealable final judgment, the parties stipulated to the dismissal of the 27 claims related to these 26 remaining videos, allowing Plaintiffs to reinstitute these claims if the district court’s ruling is vacated or reversed on any of the claims on appeal.8 The final judgment was issued on November 1, 2021, dismissing all of Plaintiffs’ other claims with prejudice. Plaintiffs filed a timely notice of appeal on November 29, 2021. STANDARD OF REVIEW The court of appeals reviews a district court’s grant of summary judgment de novo, “construing the evidence in the light most favorable to the party against whom summary judgment was granted and drawing all reasonable inferences in that party’s favor.” Bey v. City of New York, 999 F.3d 157, 164 (2d Cir. 2021) (citing Sista v. CDC Ixis N. Am., Inc., 445 F.3d 161, 168-69 (2d Cir. 2006)). When parties cross-move for summary judgment, each motion is analyzed separately, “in each case construing the evidence in the light most favorable to the non-moving party.” Schwebel v. Crandall, 967 F.3d 96, 102 (2d Cir. 2020). Summary judgment is granted only if “there is no genuine dispute as to any material fact and the movant is entitled to judgment as a matter of law.” FED. R. CIV. P. 56(a). Accordingly, a genuine dispute as to a material fact precludes summary judgment “where the evidence is such that a reasonable jury could decide in the non-movant’s favor.” Lucente v. Cnty. of Suffolk, 980 F.3d 284, 296 (2d Cir. 2020) (quoting Beyer v. Cnty. of Nassau, 524 F.3d 160, 163 (2d Cir. 2008)). DISCUSSION Plaintiffs contend that Vimeo is not entitled to safe harbor under Section 512(c)(1) of the DMCA because Vimeo cannot meet two of the statute’s requirements. They argue that Vimeo (1) had “aware[ness] of facts or circumstances from which infringing activity is apparent” (often referred to as “red flag knowledge”) and failed to “expeditiously [] remove, or disable access to, the material,” 17 U.S.C. §512(c)(1)(A)(ii)-(iii); and that it (2) “receive[d] a financial benefit directly attributable to the infringing activity [while having]…the right and ability to control such activity,” id. §512(c)(1)(B). Although it is the defendant’s burden to show that it meets the qualifications for entitlement to the safe harbor — such as by showing that it is a service provider within the meaning of the statute — we have held, citing the Nimmer copyright treatise, that it is the plaintiff’s burden to demonstrate that a service provider has lost entitlement to the safe harbor because it had actual or red flag knowledge of the infringement. See Vimeo I, 826 F.3d at 94-95. In our view, for the same reasons, a plaintiff must also bear the burden of persuasion in showing that the defendant was disqualified from the safe harbor because it received a financial benefit directly attributable to the infringing activity while having the right and ability to control such activity. See 17 U.S.C. §512(c)(1)(B). For the reasons explained below, we reject Plaintiffs’ arguments as to both bases of disqualification and hold that Plaintiffs have failed to establish a basis to deny Vimeo access to the safe harbor. 1. Red Flag Knowledge Among the conditions that the DMCA establishes for a service provider to qualify for the protection of its safe harbor are the following: that it “(i) does not have actual knowledge that the material…on the system or network is infringing; (ii) in the absence of such actual knowledge, [it] is not aware of facts or circumstances from which infringing activity is apparent [red flag knowledge]; or (iii) upon obtaining such knowledge or awareness, [it] acts expeditiously to remove, or disable access to, the material.” 17 U.S.C. §512(c)(1)(A)(i)-(iii). In Viacom, our court explained the difference between the “actual” and “red flag” knowledge provisions as follows: The difference between actual and red flag knowledge is…not between specific and generalized knowledge, but instead between a subjective and an objective standard. In other words, the actual knowledge provision turns on whether the provider actually or ‘subjectively’ knew of specific infringement, while the red flag provision turns on whether the provider was subjectively aware of facts that would have made the specific infringement ‘objectively’ obvious to a reasonable person. 676 F.3d at 31 (emphasis added). In Vimeo I, we further clarified the “reasonable person” standard relevant to the red flag knowledge analysis: “[t]he hypothetical ‘reasonable person’ to whom infringement must be obvious is an ordinary person — not endowed with specialized knowledge or expertise concerning music or the laws of copyright.” 826 F.3d at 93-94.9 We noted that under this standard, a service provider’s non-expert employees cannot be expected to necessarily know whether a particular use of copyrighted music in a video constituted infringement, or, alternatively, whether it was a fair use, see 17 U.S.C. §107, or authorized under a license. See Vimeo I, 826 F.3d at 96-97. Accordingly, the mere fact that a video contains all or substantially all of a piece of recognizable, or even famous, copyrighted music and was to some extent viewed (or even viewed in its entirety) by some employee of a service provider would be insufficient (without more) to sustain the copyright owner’s burden of showing red flag knowledge. Id. at 97. We concluded that: Vimeo is entitled to summary judgment on those videos as to the red flag knowledge issue, unless plaintiffs can point to evidence sufficient to carry their burden of proving that Vimeo personnel…knew facts making th[e] conclusion [that a video was infringing] obvious to an ordinary person who had no specialized knowledge of music or the laws of copyright. Id. at 98. However, we also acknowledged in Vimeo I that it is “entirely possible that an employee of the service provider who viewed a video did have expertise or knowledge with respect to the market for music and the laws of copyright.” Id. at 97 (emphasis added). Thus, as an alternative way to establish red flag knowledge, a plaintiff could produce evidence to demonstrate that an employee (1) was not an “ordinary person” unfamiliar with these fields, and (2) was aware of facts that would make infringement objectively obvious to a person possessing such specialized knowledge. See id. We noted, though, that “[e]ven an employee who was a copyright expert cannot be expected to know when use of a copyrighted song has been licensed,” id., and, as discussed below, even a copyright expert may similarly struggle to identify instances of fair use. Thus, in order to carry their burden of demonstrating that Vimeo had actual or red flag knowledge of the specific instances of infringement, Plaintiffs needed to show that Vimeo employees were aware of facts making it obvious to (a) a person who has no specialized knowledge or (b) a person that Plaintiffs have demonstrated does possess specialized knowledge that: (1) the videos contained copyrighted music; (2) the use of the music was not licensed; and (3) the use did not constitute fair use. Vimeo has not contested that its employees were aware that the Videosin-Suit contained copyrighted music. In support of the contention that Vimeo employees had red flag knowledge that the users were not authorized to reproduce the copyrighted music, Plaintiffs rely on evidence that Vimeo, in opening its licensing store in 2011, published a blog post saying that Vimeo employees were aware that licensing can be confusing and painful. Plaintiffs argue that if Vimeo employees knew that music licensing could be confusing and painful, then it would have been obvious to those employees that the videos they observed containing what they knew to be copyrighted music had not been licensed. To show that the employees had access to facts that made it objectively obvious the videos were not fair use, Plaintiffs rely on evidence that Vimeo told users that using copyrighted music in a video “generally (but not always) constitutes copyright infringement” and that Vimeo employees had been told by Vimeo’s legal counsel not to use copyrighted music in the background of videos they created. See Appellants’ Br. at 59 (quoting J. App’x at 904). We are not persuaded by these arguments. The fact that licensing music, as a general matter, can be challenging or confusing does not make it obvious that music accompanying a particular user-uploaded video was not licensed. Even if a person without specialized knowledge would have intuited a likelihood that many of the posted videos were not authorized, that would not make it obvious that a particular video lacked authorization to use the music. This is all the more true in view of the uncontested fact that, since 2011, Vimeo had run a store from which users could purchase licenses to use music in videos. Accordingly, Vimeo employees were aware of the existence of simplified opportunities available to purchase licenses. Furthermore, because Plaintiffs have not proved that Vimeo employees had specialized knowledge of the music industry, those employees’ awareness that music found on their videos was under copyright did not show that they knew whether the music they heard on user videos came from EMI or another label. Plaintiffs’ evidence does not support it being apparent to Vimeo employees that the music they heard on any particular video came from a label that did not offer licenses through Vimeo’s store or otherwise. Plaintiffs also rely on the contention that EMI’s cease-and-desist letter, sent to Vimeo in 2008, put Vimeo employees on notice that any EMI music used on the website was unauthorized. Plaintiffs cite EMI Christian Music Grp., Inc. v. MP3tunes, LLC, 844 F.3d 79, 93 (2d Cir. 2016), where we explained that the defendant’s subjective awareness that there had been no legal online distribution of Beatles songs could support red flag knowledge that any online electronic copies of Beatles songs on defendant’s servers were unlicensed. But the same logic does not necessarily apply here. As the district court pointed out, an awareness that EMI sent a letter in the past demanding removal of its music gave no assurance that EMI did not thereafter make contracts licensing the use of its music, especially in view of evidence that some users who posted the videos containing EMI music asserted that EMI had provided them with authorization to use the music. The DMCA does not require service providers to perform research on mere suspicion of a user’s infringement to determine the identity of the music in the user’s video, identify its source, and determine whether the user acquired a license. See Vimeo I, 826 F.3d at 98-99 (explaining, in the context of a contention of willful blindness, that requiring service providers “constantly to take stock of all information their employees may have acquired that might suggest the presence of infringements in user postings, and to undertake monitoring investigations whenever some level of suspicion was surpassed,…would largely undo the value of §512(m)”). Even if we concluded that Vimeo had red flag knowledge that EMI’s music in user videos was not authorized or licensed, that would be insufficient to satisfy Plaintiffs’ burden. Plaintiffs needed in addition to show that it would be apparent to a person without specialized knowledge of copyright law, or, alternatively, persons who have been demonstrated to possess specialized knowledge of copyright law, that the particular use of the music in the Videos-in-Suit was not fair use. Plaintiffs contend that they showed that the Vimeo staff had “legal acumen” as to copyright laws. See Appellants’ Br. at 59. We disagree. Their argument rests solely on Vimeo’s having told its employees not to produce videos containing copyrighted music and Vimeo’s having communicated to users that using copyrighted music “generally (but not always) constitutes copyright infringement.” Id. Those facts do not support the conclusion that a Vimeo employee, absent familiarity with copyright laws, would have a basis for knowing whether the use of copyrighted music in a particular video was or was not a fair use. Plaintiffs’ argument goes too far; it would require Vimeo employees to assume that uses of copyrighted material are never fair use. Vimeo’s exercise of prudence in instructing employees not to use copyrighted music and advising users that use of copyrighted music “generally (but not always) constitutes copyright infringement” did not educate its employees about how to distinguish between infringing uses and fair use. Furthermore, at least during the period in question, the boundaries of fair use were not so well settled as to make clear even to persons well acquainted with copyright law whether and when a dancing, acting, or lipdubbing performance of copyrighted music might pass muster as a fair use. The difficulty distinguishing fair use from infringement at the time in question is illustrated by the Supreme Court’s subsequent consideration of Andy Warhol Foundation for the Visual Arts, Inc. v. Goldsmith, 598 U.S. 508 (2023). The question presented to the Court in that case was whether the first statutory factor for a finding of fair use — “the purpose and character of the use,” 17 U.S.C. §107(1) — favored a finding of fair use for a portrait of the singer Prince, created by Andy Warhol in 1984, which incorporated a copyright-protected photograph of Prince taken by photographer Lynn Goldsmith, while making changes to the original. See Warhol, 598 U.S. at 516-18. More than 40 copyright experts, as “Copyright Law Professors” and “Art Law Professors,” filed amicus briefs in their own names supporting a finding of fair use and consequently no infringement.10 At the same time, approximately 18 intellectual property professors filed amicus briefs arguing that the copying did not represent fair use.11 The Court itself also proved to be bitterly divided. In the majority opinion, seven Justices rejected the arguments passionately advanced by two dissenters that Warhol’s changes to the original Goldsmith photograph were transformative and that the ultimate creation was a fair use. See Warhol, 598 U.S. at 548-50. Where academic scholars specialized in the study of the fair use question and the Justices of the Supreme Court are so divided, we cannot conclude that it was “apparent,” as required by Section 512(c)(1)(A)(ii), to untutored employees of Vimeo that dancing, acting, or lip-dubbing performances of copyrighted music uses posted by Vimeo users were not fair use. In making this observation, we express no views on the strength or weakness of arguments that such lip-dubs or dances or acting performances qualify as fair use. We are merely pointing out the weakness of Plaintiffs’ argument that Vimeo had red flag knowledge of infringement based on mere observation of the videos by employees with no training in copyright law. We reject Plaintiffs’ arguments that Vimeo lost the protection of the safe harbor by virtue of having red flag knowledge that user postings were infringing and then failing to remove those postings from its website. 2. Right and Ability to Control Plaintiffs also contend that Vimeo lost the protection of the Section 512(c) safe harbor because it “receive[d] a financial benefit directly attributable to the infringing activity, in a case in which [it had] the right and ability to control such activity.” 17 U.S.C. §512(c)(1)(B) (emphasis added). The difficulty we face at the outset in considering these questions derives from uncertainty regarding what Congress meant by “right and ability to control [the infringing] activity.” Id. Exercise of control could mean many different things. What sort of control did Congress have in mind? How much control is required? If it meant simply the legal right and the technical capability to remove videos from the site, or prevent their installation, it would be rare for a service provider not to fall within that description. In virtually all cases, private operators of websites that host material posted by users have the legal right to select the categories of videos they will allow, and to exclude those that do not conform, as well as the technical ability to effectuate these choices. Indeed, the Act’s delineation of the scope of its safe harbor presupposes that the service provider will have the “right and ability” to remove infringing material from the site, as it provides in Sections 512(c)(1)(A)(iii) & (C) that such removal is the means by which the service provider secures entitlement to the safe harbor upon becoming aware of the infringing nature of a user’s posting. See Viacom, 676 F.3d at 37. Construing Section 512(c)(1) to mean that profiting from possession of a capability that virtually all private service providers are expected to possess would effectively foreclose access to the Act’s safe harbor and would substantially undermine what has generally been understood to be one of Congress’s major objectives in passing the DMCA: encouraging entrepreneurs to establish websites that can offer the public rapid, efficient, and inexpensive means of communication by shielding service providers from liability for infringements placed on the sites by users. It seems highly unlikely that Congress intended that this ambiguous provision should be interpreted to have a meaning that would effectively undo a major benefit that the Act appears intended to confer. Accordingly, we held in Viacom that a showing of a “right and ability to control” requires “something more” than the mere ability to remove or block access to materials on its website. See 676 F.3d at 38 (internal citation omitted). In describing what the “something more” might be, we noted that (at that time) only one decision of a federal court — Perfect 10, Inc. v. Cybernet Ventures, Inc., 213 F. Supp. 2d 1146 (C.D. Cal. 2002) (“Cybernet”) — had concluded that a service provider had the right and ability to control infringing activity under Section 512(c)(1)(B).12 See Viacom, 676 F.3d at 38. In that unique instance, the service provider had demanded that its users comply with a particular layout and appearance, gave its users extensive advice on content, and engaged in stringent monitoring of images. See Cybernet, 213 F. Supp. 2d at 1173. We characterized Cybernet as an instance where a service provider exerted “substantial influence” on user activities. See Viacom, 676 F.3d at 38. We also suggested in Viacom that a service provider might have a right and ability to control infringing activity if it had induced the infringing activity. See id. (citing Metro-Goldwyn-Mayer Studios Inc. v. Grokster, Ltd., 545 U.S. 913 (2005)). Since then, only one additional federal court decision — Mavrix Photographs, LLC v. LiveJournal, Inc., 873 F.3d 1045, 1059 (9th Cir. 2017) — has acknowledged a possibility that a service provider might have come within the statutory standard of “right and ability to control.” In that case, the defendant website operator reviewed each user submission for content before allowing publication and rejected nearly two-thirds of submitted posts. See id. The Ninth Circuit remanded to the district court with instructions to assess whether those facts would justify a finding that the service provider had the “right and ability to control” infringing activity. See id. As with Cybernet, the site operator’s activities in Mavrix arguably involved exercise of “substantial influence” over user postings. Our court has addressed this issue twice since Viacom, in each case employing a non-precedential summary order to affirm a district court ruling that found no right and ability to control. First, the district court in Wolk v. Kodak Imaging Network, Inc. had ruled that the service provider did not have the right and ability to control because it neither engaged in prescreening, nor extensively influenced users regarding content, nor altered user content. See 840 F. Supp. 2d at 748. This court affirmed the judgment “substantially for the reasons stated by the district court” the judgment was proper, finding that the operator lacked the “right and ability to control” infringing activity. See Wolk v. Photobucket.com, Inc., 569 F. App’x 51, 52 (2d Cir. 2014) (summary order). Second, the district court in Obodai v. Demand Media, Inc. had ruled that evidence that the service provider monitored site traffic, but not the content of user postings, was insufficient to “conclude that [the] defendant exercised control over user submissions sufficient to remove it from the safe harbor provision of section 512(c)(1)(B).” No. 11-cv-2503 (PKC), 2012 WL 2189740, at *8 (S.D.N.Y. June 13, 2012). As in Wolk, this court found that “the district court correctly determined that [the defendant] was eligible for” the Section 512(c) safe harbor. See Obodai v. Cracked Ent. Inc., 522 F. App’x 41, 42 (2d Cir. 2013) (summary order). Plaintiffs endorse the exercise of “substantial influence” over infringing activity as the governing standard. See Appellants’ Br. at 27 (“[The Act] preserves vicarious liability for providers…who can exercise ‘substantial influence’ over users’ infringing activity, and who derive a financial benefit from that activity. Viacom, 676 F.3d at 38 (citing Cybernet, 213 F. Supp. 2d at 1173).”). Plaintiffs argue that a service “provider’s influence should be found ‘substantial’ when the provider exercises editorial judgment, such as by evaluating content for its merit.” Appellants’ Br. at 28. They contend that, by promoting and demoting user posts based on their merit or their appeal to other users, as well as by banning videos that merely reproduced pre-existing content and gameplay videos, Vimeo exercised the sort of substantial influence that, combined with the profits it earned, at least raised a jury question of its eligibility for the safe harbor, thus requiring that we vacate the district court’s grant of summary judgment to Vimeo. They assert that “Vimeo employees made significant editorial judgments about the precise kind of activity — user uploads — that infringe copyright” and that “Vimeo’s level of control over its users’ activity outstripped” the controls found in Cybernet to constitute exercise of “substantial influence” over the user postings, resulting in forfeiture of access to the safe harbor. See Appellants’ Br. at 35. We cannot agree. In the first place, the extent to which Vimeo employees exercised control over user posts was far less intrusive than the controls exercised by the service providers in Cybernet and Mavrix. As noted above, in Cybernet, the service provider dictated a prescribed layout and appearance for all user posts, gave its users extensive advice on content, and engaged in stringent monitoring of images. See 213 F. Supp. 2d at 1173. In Mavrix, the service provider reviewed each user submission for content before allowing publication and rejected nearly two-thirds of submitted posts. See 873 F.3d at 1059. Vimeo’s intrusions into user autonomy over their posts were far less extensive as to both coercive effect and frequency. Calling attention to selected videos by giving them a sign of approval or displaying them on a Staff Picks channel (or the contrary, by demoting them) did not restrict the freedom of users to post whatever videos they wished. As for Vimeo’s insistence that user videos be limited to those created at least in part by the user, and its ban of pornography as well as gameplay videos and other unoriginal content, this was somewhat more intrusive. But these requirements were in the nature of (i) avoiding illegality and the risk of offending viewers and (ii) designing a website that would be appealing to users with particular interests. It seems unlikely that, in Congress’s use of an ambiguous term in formulating the standards for eligibility for the safe harbor to encourage entrepreneurs to create websites, it intended to deny eligibility for the safe harbor to entrepreneurs merely because they sought to exclude content that violates other laws or because they sought to design sites to make them appealing to selected categories of consumer preferences — whether for child-friendly videos, videos devoted to dance, kittens and puppies, hunting and fishing, cars, baseball, wildlife, antiques, carpentry, or whatever else. The creation of websites designed to satisfy consumer demands appears to be precisely the sort of entrepreneurial activity that the safe harbor was intended to encourage.13 In addition, when one recognizes the huge number of videos posted by users on Vimeo, Plaintiffs have failed to show that interventions by Vimeo staff affected more than a tiny percentage. For example, in 2012, 43,000 new videos per day were posted on Vimeo, which annualized to over 15 million new videos. During that year, Vimeo had only 74 employees. Apart from how minimal an intrusion it is for a staff member to select a video to receive an indication of approval, the number of videos that 74 staff members could have evaluated and emphasized amounted to no more than an insignificant percentage of those posted. Plaintiffs did not show that staff awards, consisting of likes, thumbs-ups, and promotions to a Staff Picks channel (or demotions), came anywhere near amounting to exercising “substantial influence” over the contents of user-posted videos. In our view, denial of eligibility for the safe harbor based on such noncoercive exercises of control over only a small percentage of postings would undermine, rather than carry out, Congress’s purposes in establishing the safe harbor. In establishing this safe harbor with its limitations, Congress sought to achieve a compromise with the following complex objectives. First, Congress recognized that the creation of websites on which the public could post videos would render a hugely valuable public service. However, the expense of either policing all postings to weed out infringements or of paying damages for infringements by users would be prohibitive. Entrepreneurs could not be expected to establish such ventures if doing so would expose them to an open-ended risk of liability for the posting by users of infringing videos or if they would need to incur unsustainable costs in policing posted videos to ensure that they were free of infringements. See S. REP. No. 105-190, at 8 (1998) (“[W]ithout clarification of their liability, service providers may hesitate to make the necessary investment in the expansion of the speed and capacity of the Internet.”). Congress therefore enacted inducements to establish such websites by granting safe harbors protecting service providers from liability for infringements posted by users and by expressly exempting the service providers from any obligation to conduct burdensome research to detect infringements. See 17 U.S.C. §512(m) (“Nothing in this section shall be construed to [require] a service provider [to] monitor[] its service or affirmatively seek[] facts indicating infringing activity….”). At the same time, Congress recognized that the posting of infringements by users of websites could cause significant economic harm to copyright holders. Accordingly, Congress placed some limitations on eligibility for the safe harbor. While Congress deemed it important not to impose on website operators the huge burden of checking user posts for infringement, it recognized that this burden would be considerably lessened if the operator was already voluntarily incurring a large expense in monitoring and controlling user posts to serve the operator’s own business purposes. In such cases, where an operator is not merely passively accepting content but is arguably playing a large role in shaping the content of user posts, checking also for infringements would add only a relatively modest incremental expense and would not substantially disincentivize the provision of socially valuable sites. Congress therefore gave rightsholders some limited recourse against service providers that have the “right and ability to control” infringements by users, which our court has interpreted to apply in circumstances when the service provider has exercised “substantial influence” over user activities.14 To interpret this provision as Plaintiffs argue — to deny Vimeo access to the safe harbor merely because of the tiny influences it exercised — would subject Vimeo to a huge expense in monitoring millions of posts to protect itself against the possibility of liability for infringements. It would undermine the compromise that we understand Congress to have sought. It would prevent service providers from seeking to make their websites responsive to user desires, substantially diminishing their utility to the public. We recognize, as Congress did, that this compromise will cause some hardships to rightsholders. At the same time, increasing the vulnerability of service providers to liability for infringements posted by users would result in diminishing the scope and availability of web services offered by service providers to the public for lawful use. Where the balance should ideally be struck is a policy question committed to the judgment of Congress. If Congress believes we have misunderstood its compromise or has changed its mind as to where the balance should lie, it is for Congress to pass corrective legislation. Last, we address the question whether by promoting the creation of lipdub videos, given the likelihood that users would lip-dub currently popular copyrighted music, Vimeo encouraged infringement and should be deemed to have thus exercised the “right and ability to control,” risking forfeiture of the safe harbor. We see force in the argument that encouraging users to make infringing lip-dubs should trigger forfeiture of a safe harbor designed to protect service providers from liability for infringements for which they were in no way responsible. However, Plaintiffs have waived the argument. Their opening brief declines to assert this argument, explaining that our earlier opinion in this case “forecloses (at this stage) the argument that Vimeo’s ‘urging’ and ‘encouraging users to post infringing material’ constituted inducement under Grokster.” Appellants’ Br. at 28 n.5 (quoting Vimeo I, 826 F.3d at 99).15 We do not read our opinion in Vimeo I as foreclosing this potentially forceful argument. Our comments in Vimeo I to the effect that Plaintiffs’ arguments were not supported by the evidence concerned a different issue: willful blindness to infringement, which we ruled could not be demonstrated by “a handful of sporadic instances…in which Vimeo employees inappropriately encouraged users to post videos that infringed music.” Vimeo I, 826 F.3d at 99. In Vimeo I, we rejected Plaintiffs’ contention of Vimeo’s willful blindness in substantial part because of the tiny scope of isolated instances of a different sort of encouragement to infringe: where employees deviated from company policy by telling users not to worry about infringement.16 Our opinion neither said nor implied that encouragements to infringe could not impact Vimeo’s entitlement to the safe harbor under Section 512(c)(1)(B). The discussion furthermore did not concern Vimeo’s policy to encourage lip-dubs. We can see no basis for Plaintiffs’ reading our Vimeo I opinion as “foreclos[ing]…the argument that Vimeo’s ‘urging’ and ‘encouraging users to post infringing material’ constituted inducement.” Appellants’ Br. at 28 n.5. Insofar as this concerns the resolution of this appeal, the answer is simple. Because Plaintiffs waived the argument that Vimeo’s policy to encourage lip-dub videos constituted (or contributed to) the “right and ability to control” infringements, with adverse consequences under Section 512(c)(1)(B), we do not consider that as a possible basis of liability for Vimeo. Because we agree with the district court that Vimeo’s other activities — commenting on and promoting posted videos, and banning certain types of videos — do not, in combination, support a finding that Vimeo has exercised “substantial influence” over the infringing activity, as required by Viacom, Plaintiffs failed to raise a triable question regarding Vimeo’s “right and ability to control.” Accordingly, we need not reach the question whether Vimeo received a financial benefit directly attributable to that infringing material. CONCLUSION Plaintiffs failed to demonstrate a disputed issue of material fact as to whether Vimeo had the right and ability to control infringing activity on its website or whether Vimeo’s employees had red flag knowledge of users’ infringement. Accordingly, we AFFIRM the judgment of the district court.
