CASRIP Newsletter - Summer 2007, Volume 14, Issue 3
What Does YouTube Know?
The Digital Millennium Copyright Act Section 512(c)(1)(A) and the burden to police a website for copyright violations
By Samantha Schmidt[*]
I. INTRODUCTION
The Internet has not only revolutionized our ability to access and disseminate information but has also shaken the foundations of copyright law. “[T]he Internet turns every user in every living room into a mass distributor of just about anything that can be digitized, including film, photography, the written word and, of course, music.” [1] Thus, the Internet has forced a “radical rethinking of…copyright law.” [2] In 1998, Congress passed the Digital Millennium Copyright Act (“DMCA”) and in doing so attempted to strike the appropriate balance between the competing interests of content owners of copyrighted works, Internet service providers and end users. But, at the time the DMCA was crafted, Congress was focused on “promoting electronic commerce” as well as ensuring Internet service providers had an incentive to continue investing in the expansion of the Internet’s infrastructure.[3] As a result, Congress did not anticipate Web 2.0 business models and the proliferation of creative content on “user-generated content” websites.[4]
Web 2.0 “user-generated content” websites differ from the traditional online service providers in two main respects. First, the popularity of Web 2.0 “user-generated content” websites is largely tied to content and the sharing of content. In contrast, the services offered by traditional Internet service providers are related to Internet connectivity, e-commerce sites, and the access and exchange of information. Second, “user-generated content” websites are not merely passive conduits like the more traditional online service providers. Rather, Web 2.0 companies have built websites specifically to allow users to share creative content (both original and copyrighted). The fact that the popularity of “user-generated content” websites is tied to creative content and the sharing of creative content has resulted in contentious battles between the sites and creative content owners.
In particular, content owners argue that Web 2.0 service providers are profiting from unauthorized copyrighted content and should be liable to the content owners for copyright infringement. Content owners also argue that they should not bear the sole burden for policing websites for copyrighted material. On the other hand, “user-generated content” service providers claim that they are protected by the DMCA section 512(c) safe harbor, which limits the liability of service providers for copyright infringement due to the storage of copyrighted material on a service provider’s network at the direction of a user. And, further, because they believe that they are protected by the safe harbor, Web 2.0 service providers claim they have no obligation to monitor their sites for infringing material.
The battle between “user-generated content” service providers and content owners hinges on the interpretation of DMCA section 512(c)(1)(A). This section requires service providers to remove or block access to infringing material if the service provider has actual knowledge of infringing material or apparent knowledge of infringing activity on its network or system. As a result, a service provider with actual or apparent knowledge of infringing activity taking place on its site may bear the burden to police its site in order to remain eligible for the section 512(c) safe harbor. This specific issue has taken center stage in the Viacom v. YouTube lawsuit.[5] This specific issue, however, has not yet been addressed by the courts, and remains a subject of legal uncertainty.
As a result, this paper explores the issue of whether, in order to be eligible for the section 512(c) safe harbor, YouTube has an obligation to monitor its website for infringing material. Part II of the paper reviews the history and background of the Digital Millennium Copyright Act, including a discussion of the “Information Task Force” and the Lehman Working Group’s role in crafting the legislation. Part III examines the Web 2.0 “user-generated content” websites and, in particular, YouTube.com. In addition, Part III discusses the underlying issues in the Viacom v. YouTube lawsuit. Part IV analyzes DMCA section 512(c) and, in particular, section 512(c)(1)(A) and the relevant subsections in relation to the Viacom v. YouTube lawsuit. Specifically, Part IV examines the legislative history of section 512(c)(1)(A) and case law to determine the meaning of “actual knowledge” and “apparent knowledge” as those terms are used in 512(c)(1)(A). The last section of Part IV discusses whether YouTube has actual knowledge of infringing material or apparent knowledge of infringing activity on its site and explores the resulting implications. Part V of the paper addresses the issue of monitoring under the DMCA and the unintended consequences of monitoring a website. Last, in Part VI this paper suggests that YouTube should monitor its website for infringing material in order to satisfy the section 512(c)(1)(A) criteria and remain eligible for the section 512(c) safe harbor. This paper also suggests that YouTube will likely not suffer any unintended consequences for monitoring its website for infringing content.
II. THE DIGITAL MILLENNIUM COPYRIGHT ACT: PRESERVING THE NII
A. Background
The Digital Millennium Copyright Act (hereinafter, “DMCA” or “Act”) was passed by Congress and signed into law by President Clinton on October 28, 1998.[6] The purpose of the DMCA “is to facilitate the robust development and world-wide expansion of electronic commerce, communications, research, development and education in the digital age.”[7] Specifically, the DMCA implemented two World Intellectual Property Organization (WIPO) treaties related to the technological protection of copyrighted works.[8]
The DMCA also addressed another significant issue in Title II of the Act: the “Online Copyright Infringement Liability Limitation Act (section 512 of the Copyright Act) – which limits the direct and indirect liability of online and Internet service providers through the establishment of various safe harbors.[9] Prior to the passage of Title II, there was no agreement amongst the courts regarding online service providers’ liability for copyright infringement.[10] Due to the uncertainty for both service providers and content owners, Title II represented a compromise between these groups;[11] that is, Title II tried to strike a balance between promoting the continued economic growth and development of the Internet’s infrastructure, while simultaneously protecting intellectual property rights.[12]
B. The Lehman Working Group
The passage of the DMCA was a five-year process and a result of the culmination of many factors.[13] Specifically, shortly after President Clinton took office in 1993, the “Information Infrastructure Task Force” was formed for the sole purpose of devising government policy regarding the Internet.[14] A subgroup of the task force, the “Working Group on Intellectual Property,” was appointed and Bruce Lehman, the Patent Commissioner, was appointed to chair the group.[15] The purpose of the Working Group was to “investigate the effects of emerging digital technology on intellectual property rights and make recommendations on any appropriate changes to U.S. intellectual property law and policy.”[16]
The Working Group held public hearings and released its preliminary findings (the “Green Paper”), all of which were incorporated into its final assessment (the “White Paper”).[17] Ultimately, the Working Group concluded that copyright owners should get enhanced copyright protection in relation to the Internet because otherwise, content owners “will not be willing to put their interests at risk if appropriate systems…are not in place to permit them to set and enforce the terms and conditions under which their works are made available” on the Internet.[18] In other words, the Lehman Working Group specifically tied the success of the Internet to content.[19] As a result, the Lehman Working Group concluded, among other things, that copyright owners should be able to use technical protections to protect their works. In addition, any transmissions of copyrighted works on the Internet at the direction of end users constituted copyright infringement.[20] Thus, the Lehman Working Group determined that online service providers should be liable for direct, contributory, and vicarious infringement as a result of the activities of their users. [21]
C. The DMCA: Early Legislation
Soon after the release of the Lehman Working Group’s White Paper in 1995, both the House of Representatives and the Senate introduced legislation to address the online service provider liability issue raised by the group.[22] However, various interest groups (such as online service providers, libraries, telephone companies, consumer electronics manufacturers, law professors, Internet civil liberties groups and computer hardware manufactures) banded together to defeat the proposed legislation.[23] The interest groups were concerned with the Lehman Working Group’s anti-circumvention proposals as well as the Group’s suggestions related to the liability of online service providers.[24] As a result of interest group efforts, the proposed legislation was tabled for a few years until after the WIPO treaties were agreed upon by all WIPO member countries.
Three years later, two bills were introduced in the House of Representatives and the Senate to implement the WIPO treaties as well as address the issue of online service provider liability.[25] In fact, several interest groups (including telephone companies, commercial Internet service providers, schools and libraries) “insisted that an agreement setting up a safe harbor for online service providers was a precondition to the enactment of the [WIPO] implementing legislation.”[26] As a result, Title II of the Act (the online service provider safe harbor provisions) represents a long-fought but mutual agreement between the lobbyists for the entertainment industry (content owners) and the lobbyists for the Bells (the telecommunication companies and owners of the Internet’s infrastructure).[27]
Specifically, “[c]ontent owners agreed that Internet service providers should not be liable for their subscribers’ infringing transmissions so long as the provider had no reason to suspect infringement was taking place, on the condition that the service provider agreed to shut down copyright violators and remove infringing material as soon as a content owner notified it of a violation.”[28] The Act created “’strong incentives for service providers and copyright owners to cooperate to detect and deal with copyright infringements that take place in the digital network environment.’”[29] Title II also represents Congress’ attempt to balance the competing interests between copyright owners and end users.[30] Specifically, the DMCA strikes a balance between the need for content owners to have infringing works quickly taken down and end users’ interests in not having their material removed from a website without recourse.[31]
D. The DMCA: Safe Harbor Provisions
Title II, codified at section 512 of the Copyright Act, limits the liability of service providers for copyright infringement through the creation of four safe harbors. Specifically, the safe harbor provisions provide “a complete bar on monetary damages, and restrict the availability of injunctive relief”[32] for direct, vicarious and contributory infringement.[33] However, in order for a service provider to enjoy any one of the safe harbor provisions, it must first satisfy several threshold requirements.[34]
First, the party seeking protection under a safe harbor must establish that they meet the definition of a “service provider” as defined by section 512(k)(1)(A) or section 512(k)(1)(B).[35] Second, a service provider must also show that it has “adopted and reasonably implemented, and informs subscribers and account holders of the service provider's system or network of, a policy that provides for the termination in appropriate circumstances of subscribers and account holders of the service provider's system or network who are repeat infringers; and accommodates and does not interfere with standard technical measures.”[36] Last, to be eligible for protection under section 512(c), the service provider must also “designate[] an agent to receive notifications of claimed copyright infringement.”[37]
The four safe harbor provisions correspond to the four functional operations of a service provider. In particular, the safe harbors limit the liability of service providers for the following activities: (1) §512(a): transitory digital network communications; (2) §512(b): system caching; (3) §512(c): storage of information on systems or networks at the direction of users; and (4) §512(d): information location tools.[38] The safe harbors are separate and distinct from one another.[39] Thus, if a service provider does not qualify for one of the safe harbors, the service provider is not precluded from qualifying for a different safe harbor. In addition, a service provider is not liable for copyright infringement simply because it does not qualify for any of the safe harbor provisions.[40] In other words, “[s]ection 512 does not define what is actionable copyright infringement in the online environment and does not create any new exceptions to the exclusive rights under copyright law” nor “affect any defense available to a service provider.”[41] Rather, the purpose of section 512 is to provide service providers with some certainty regarding exposure to liability for copyright infringement.[42] Last, the applicability of the safe harbors is not conditioned on a service provider monitoring or affirmatively seeking facts indicating infringing activity on its system or network.[43]
III. WEB 2.0: COMMUNITY, SHARING AND CREATIVE CONTENT
A. Web 2.0 v. the DMCA
The DMCA “was not passed with [blogs, p2p or video file] sharing services in mind….”[44] Rather, at the time the DMCA was enacted, Internet usage was limited mostly to postings to bulletin boards, chat rooms, email and electronic commerce.[45] The dominant Internet service providers were AOL, Yahoo, telephone companies and other telecommunication companies that supplied the Internet’s infrastructure and basic services as well as some early electronic commerce sites.[46] YouTube, MySpace, Facebook, Blogger, Bolt and other “user-generated content” websites had not yet been developed.[47] Thus, there is a tension regarding the applicability of the DMCA to “user-generated content” websites.
B. What are “user-generated content” websites?
“User-generated content” websites such as the video-sharing site, YouTube, and the online community site, MySpace, allow users to post and share original and copyrighted videos and content.[48] Further, users on YouTube, for example, are not only able to upload and share videos on the site but users can also “[i]ntegrate YouTube videos on websites using video embeds or APIs.”[49] The popularity of the “user-generated content” websites is tremendous. For example, on YouTube nearly 100 million clips are viewed daily and 65,000 new videos are uploaded every 24 hours.[50] In addition, the site has almost 20 million visitors each month and it already commands up to 64% of the UK online video market.[51] However, the popularity of “user-generated content” sites is not solely due to original user-generated content posted on the sites. Rather, much of the traffic driven to these sites is due to the unauthorized posting of copyrighted material on a daily basis.[52]
C. Web 2.0 and Content Owners: the battle over copyrighted works
YouTube owes much of its early success to a user uploading a copyrighted video clip from a Saturday Night Live sketch, “Lazy Sunday,” to the site.[53] By the time NBC asked YouTube to remove the video clip, 5 million people had viewed it and YouTube had become an overnight sensation.[54] However, despite the removal of the “Lazy Sunday” video, users still continue to upload copyrighted video clips to YouTube and other “user-generated content” websites every day.[55] For example, “NBC Universal [who entered into a licensing deal with YouTube] has three employees who troll [YouTube] every day looking for studio-owned material, and they send more than 1,000 [takedown] requests a month to YouTube.”[56] As a result of the proliferation of unauthorized copyrighted content appearing on YouTube and other “user-generated content” websites, content owners have not only sent thousands of takedown notices to these sites but they have also rattled their litigation sabers. Content owners do not appreciate the fact that unauthorized copyrighted material is being uploaded to the websites because the content owners realize no income stream from these illegal postings. Rather, the unauthorized content creates “hundreds of millions of dollars of value for the owners of MySpace,” YouTube and other community and video sharing websites.[57]
The discontent of content owners, including the threat and filing of copyright infringement lawsuits, has led some of the “user-generated content” websites to install filtering technology, while others attempt to negotiate licensing deals with content owners.[58] For example, after MySpace was sued by Universal it began to remove “Universal content uploaded by its users.”[59] In fact, now, MySpace employs filtering technology to preemptively block the ability of users to upload Universal works. In addition, some video sites, like Revver.com, prescreen every video for copyright infringement before it is made available for viewing on the site.[60] On the other hand, YouTube has stalled efforts to deploy filtering technology to detect/prevent/remove unauthorized copyrighted works, choosing instead to concentrate on negotiating licensing deals with content owners.[61] But, YouTube has started filtering content for those content companies that have agreed to forge licensing deals with them.[62] In addition, YouTube, in an effort to deter the uploading of unauthorized movies, has a ten minute maximum limit on video length.[63]
IV. VIACOM v. YOUTUBE
The battle between content owners and “user-generated content” websites recently reached a crescendo when Viacom sued YouTube for $1 billion and injunctive relief for direct and indirect copyright infringement.[64] According to Viacom, the amount of infringing activity taking place on YouTube is massive. For example, before filing the lawsuit, Viacom ordered approximately 100,000 video clips removed from YouTube, including clips from “The Daily Show with Jon Stewart,” “The Colbert Report,” and “South Park.”[65] Viacom claims the video clips were viewed 1.2 billion times on YouTube. [66] As a result of the substantial amount of Viacom’s content available on YouTube, Viacom spends more than $100,000 a month to hunt down unauthorized material on YouTube and send takedown notices to YouTube.[67]
Despite the plethora of infringing content on its site, YouTube claims it “has no idea whether material on [its site] is being posted illegally or not.”[68] Further, although YouTube admits to using filtering technology in limited circumstances, the company takes the position it has no obligation to do so under the DMCA.[69] According to YouTube, the DMCA requires it to remove infringing content from its site only after it receives a takedown notice from a copyright owner. [70]
Viacom’s frustration with YouTube is due to several factors. First, Viacom currently bears the entire burden of policing YouTube’s site for infringing material while YouTube earns ad revenue from users viewing the infringing content on its site. Second, the fact that Viacom’s works are uploaded to YouTube without Viacom’s permission means Viacom loses the ability to control the dissemination of its works.[71] Third, Viacom is unable to establish licensing deals for electronic distribution of its content when its unauthorized works are persistently present on YouTube for free.[72]
Last, Viacom believes YouTube has “chosen not to take reasonable precautions to deter the rampant infringement on its site” and has “deliberately withheld the application of” filtering technology even though it has the technical capability to install this measure.[73] Arguably, the fact that YouTube has chosen not to filter or monitor its site for infringing material is because doing so “would cause [YouTube] to lose the benefit of the vas[t] audience the YouTube Web site now enjoys and the substantial ‘draw’ that such infringing material represents in attracting users.”[74] However, some content owners are not necessarily interested in shutting down YouTube, as YouTube and other “user-generated content” sites represent an enormous marketing avenue for the content owners; reaching viewership and awareness through the online medium that many content owners have been slow to exploit.[75] Rather, content owners want compensation from YouTube for their content.
The fact YouTube will remove material only after receiving a notice of claimed infringement does not sit well with content owners. The reason is because as soon as a content owner sends a notification and YouTube removes a copyrighted video clip from its site, the same video clip is uploaded again within a matter of hours.[76] In its defense, YouTube states its efforts go beyond what is required of it under the law because it (1) “provides tools to help copyright holders instantly alert YouTube about content they want removed from the site”; (2) “respond[s] quickly to requests for removal”; and (3) “uses digital hashing technology to keep users from uploading content that has been previously removed.”[77] But for Viacom and other content owners, without more preventive and protective measures from YouTube, such as the installation of filtering technology, the contentious battle over copyright infringement is far from resolved.[78]
Thus, the conflict between Viacom and YouTube centers on YouTube’s burden, if any, to monitor its site for infringing content. Content owners want YouTube (and other “user-generated content” websites) to bear a significant burden of policing rather than be forced to file hundreds of lawsuits against individual users.[79] The issue of whether YouTube has any burden to police its site turns on the interpretation of DMCA section 512(c)(1)(A). Section 512(c)(1)(A) requires a service provider with actual knowledge of infringing material or apparent knowledge of infringing activity on its site to remove or block access to the infringing material in order to be eligible for the safe harbor provision.[80] But the section is not clear on whether the service provider’s duty to remove or block access to infringing material also imposes a burden on a service provider to monitor its site for infringing material, and, this issue has not yet been thoroughly tested by the courts.[81] Thus, whether YouTube has an obligation to police its site depends upon the interpretation of DMCA section 512(c)(1)(A).
V. SECTION 512(c)(1)(A) AND THE KNOWLEDGE STANDARD
A. Section 512(c) and subsection (c)(1)(A)
Section 512(c) limits the liability of service providers “for infringement of copyright 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.”[82] However, in order to be eligible for the section 512(c) safe harbor, a service provider must meet not only the section 512 threshold requirements but also the following conditions: “(1) the provider must not have the requisite level of knowledge of the infringing activity…; (2) if the provider has the right and ability to control the infringing activity, it must not receive a financial benefit directly attributable to the infringing activity; and (3) upon receiving proper notification of claimed infringement, the provider must expeditiously take down or block access to the material. In addition, a service provider must have filed with the Copyright Office a designation of an agent to receive notifications of claimed infringement.”[83]
For purposes of satisfying subsection 512(c)(1)(A), a service provider must meet a three-prong test. First, a service provider can “not have actual knowledge that the material or an activity using the material on [its] system or network is infringing.”[84] Second, “in the absence of such actual knowledge,” a service provider must not be “aware of facts or circumstances from which infringing activity is apparent.”[85] Last, “upon obtaining such knowledge or awareness,” a service provider is required to “expeditiously [] remove, or disable access to, the [infringing] material.”[86] But neither actual knowledge nor apparent knowledge are defined in subsection 512(c)(1)(A) or anywhere else in section 512. As a result, it is necessary to look to the legislative history of subsection 512(c)(1)(A) as well as any relevant case law to define these terms.
B. Section 512(c)(1)(A)(i): What is actual knowledge?
Section 512(c)(1)(A)(i) provides that a “service provider shall not be liable for monetary relief, or,…for injunctive relief or other equitable relief, for infringement of copyright by reason of the storage at the direction of a user…if the service provider – does not have actual knowledge that the material or an activity using the material on the system or network is infringing.”[87] Thus, in order for a service provider to qualify for protection under section 512(c)(1)(A)(i), the service provider cannot have actual knowledge that: (1) the material on the service provider’s system or network is infringing or (2) that an activity using the material on the service provider’s system or network is infringing.[88] For the purpose of section 512(c)(1)(A), activity means “using the material on the system or network” or a “wrongful activity that is occurring at the site on the provider’s system or network at which the material resides.”[89]
“Actual knowledge” is not defined in the DMCA. However, the knowledge standard articulated in section 512(c)(1)(A) “appears borrowed from contributory infringement.” [90] As a result, legislative history and contributory infringement case law may be instructive in determining the meaning of actual knowledge. Section 512(c)(3)(B)(i) may also be helpful in determining the meaning of “actual knowledge.” That section provides that a notification of claimed infringement that does not substantially comply with the requirements set forth in section 512(c)(3) may not be considered in determining whether a service provider has actual knowledge.[91] Thus, a service provider will have actual knowledge of infringing material on its site if it receives a notification of claimed infringement from a content owner that substantially complies with section 512(c)(3).[92]
Under section 512(c)(1)(A)(i), the service provider must have actual knowledge of specific infringing material or a specific activity using infringing material on its site in order for the service provider to potentially lose eligibility for the safe harbor.[93] As a result, a notification of claimed infringement, as noted above, would provide the service provider with actual knowledge of specific infringing material on its site. In addition, courts have also concluded that actual knowledge turns on a service provider having knowledge of specific acts of infringement on its site.[94] Thus, actual knowledge may be established where a service provider uses “a screening procedure in which [its] employees view[] all files in the upload file and move[] them into the generally available files for [users].”[95] Actual knowledge of specific acts of infringement would be imputed to the service provider as a result of the service provider’s employees reviewing the infringing material before it is uploaded. But, a service provider will not have actual knowledge of specific acts of infringement if there is “’at least a colorable claim’ of noninfringement.”[96] In other words, the service provider may not have actual knowledge of specific acts of infringement if the use of the unauthorized material is a de minimis or fair one.
The determination of whether a service provider has actual knowledge of specific infringement turns on the facts of each case. For example, courts have determined that the actual knowledge standard is met based on the following evidence: (1) internal documents discussing the infringing content and activity, [97] (2) deposition testimony discussing the availability and exchange of infringing content,[98] (3) cease and desist letters that included screen shots of the unauthorized works, [99] and (4) a tutorial provided by the service provider which demonstrated how to infringe copyright and included infringing material.[100]
In addition, the use of encryption technology may not necessarily shield a service provider from having actual knowledge of infringing material on its site where the infringing activity is pervasive throughout the site.[101] On the other hand, another court has held that notices provided by other copyright owners and evidence suggesting that a service provider is aware that a content owner licenses its works are not sufficient to establish actual knowledge of specific infringing material.[102]
C. Does YouTube have actual knowledge that the material or an activity using the material on its system or network is infringing?
At a minimum, YouTube has actual knowledge of infringing material or infringing activity on its site when content owners send YouTube notifications of claimed infringement related to specific infringing material. To remain within the safe harbor, YouTube must expeditiously remove or block access to the infringing material. However, without notifications of claimed infringement or any other facts (such as internal documents or testimony) that demonstrate YouTube’s level of knowledge regarding specific infringing material, the actual knowledge standard is not satisfied. As a result, it is plausible for YouTube to claim it does not have actual knowledge of specific infringing material uploaded to its site on an instant by instant basis.[103]
On the other hand, YouTube employees scour the site for video clips to “feature on the YouTube home page,” as well as remove pornography and hateful material. As a result, YouTube may have actual knowledge of specific infringing material on its site.[104] Yet, a colorable claim of non-infringement may refute any actual knowledge of specific infringing material. Thus, it is uncertain whether YouTube has actual knowledge of specific infringing material or infringing activity on its site and, therefore, satisfies the first prong under section 512(c)(1)(A) to qualify for the safe harbor.
D. Section 512(c)(1)(A)(ii): apparent knowledge of infringing activity
Although YouTube may not have actual knowledge of specific infringing material on its site, it may have apparent or constructive knowledge of infringing activity on its site and, thus, YouTube may have an obligation to remove the infringing material even before it receives notifications of claimed infringement from content owners. The apparent knowledge standard articulated in section 512(c)(1)(A)(ii) is more expansive than the actual knowledge standard in section 512(c)(1)(A)(i). Specifically, in order for a service provider to have apparent knowledge, the service provider must only be aware of facts or circumstances of infringing activity taking place on its site, not specific acts of infringement as is required for actual knowledge.[105]
The apparent knowledge standard is not defined in section 512(c)(1)(A)(ii). However, the legislative history of that section defines apparent knowledge “as a ‘red flag’ test.”[106] The “red flag” test has both a subjective and objective element to it. [107] Under the first prong of the “red flag” test, “the subjective awareness of the service provider of the facts or circumstances in question must be determined.”[108] Thus, this prong requires a content owner to show by way of the service provider’s internal documents or other evidence that the service provider was subjectively aware of infringing activity taking place on its system or network. The second prong of the “red flag” test requires an objective analysis. The objective analysis looks at whether the infringing activity would “have been apparent to a reasonable person operating under the same or similar circumstances.”[109] In other words, did the service provider know “of facts that would have prompted a reasonable person to inquire into whether an infringement was occurring” on the service provider’s system or network.[110] This prong would be satisfied if “the service provider deliberately proceeded in the face of blatant factors of which it was aware.”[111] As a result, a service provider that “turn[s] a blind eye to “red flags” of obvious infringement” satisfies the apparent knowledge standard.[112]
“Congress’s [sic] discussion of apparent knowledge, and what evidence demonstrates such knowledge, is instructive.”[113] Specifically, “use [of] words such as ‘pirate,’ ‘bootleg,’ or slang terms in their URL and header information to make their illegal purpose obvious,” according to Congress, satisfy the ‘red flag’ test. [114] This is because the infringing nature of the websites is apparent to a reasonable person. [115] In addition, a brief or casual viewing of a website that reveals the infringing nature of the site would also be a “’red flag’ of obvious infringement.[116] As a result, a service provider aware of such “red flags” would have apparent knowledge of the infringing activity.[117] However, “[a]bsent such ’red flags’ or actual knowledge, a [service] provider would not be… aware [of infringing activity] merely because it saw one or more well known [copyrighted works] at a site….”[118]
The mere fact a website does not use terms like “pirate” or “bootleg” does not necessarily preclude a finding of apparent knowledge. Rather, the copyright owner need only show “that the provider was aware of facts from which infringing activity was apparent.”[119] But, apparent or constructive knowledge may not be imputed “merely because…[a] technology may be used to infringe [a content owner’s] copyrights.”[120] Further, general knowledge that infringing activity may be taking place is not sufficient to satisfy the apparent knowledge standard.[121]
For example, in Corbis v. Amazon, Corbis sued Amazon for copyright infringement because vendors were selling posters of Corbis’ copyrighted works on an Amazon “Marketplace” platform called “zShops.”[122] Amazon claimed it was protected from liability for copyright infringement by DMCA section 512(c) because it had no actual or apparent knowledge of infringing material on its zShops platform. [123] And, Corbis failed to produce any evidence sufficient to show that Amazon had actual or constructive knowledge of the infringing activity.[124]
Specifically, Corbis did not send a notification of claimed infringement to Amazon but rather filed a complaint. In addition, the mere fact that Amazon’s representatives met with Corbis’ representatives and Amazon’s corporate counsel was employed at Corbis were also not sufficient to establish the requisite level of knowledge. Further, Amazon’s receipt of notices about non-Corbis related infringements were also not “red flags” demonstrating that Amazon had apparent knowledge of Corbis’ infringing material on its zShops platform. Rather, the evidence “only suggest[ed] that Amazon had general knowledge that the photos may be the subject of online copyright infringement” and, as such, the evidence was not sufficient to “infer that Amazon was aware of, but chose to ignore, red flags of blatant copyright infringement on specific zShops sites.”[125] As a result, the apparent knowledge standard under the DMCA requires that “information…conclusively demonstrat[e], that [users] committed infringement.”[126]
In contrast, in the case of In re Amister, evidence showing that users openly discussed trafficking in copyrighted material and ‘screwing’ the RIAA in a service provider’s chat rooms and on bulletin boards systems, as well as a special club for users that listed the top hits on the website which included copyrighted works are examples of “red flags” of blatant infringement.[127] As such, these “red flags” were sufficient for a court to conclude that the service provider had apparent knowledge of infringing activity.[128] In addition, a service provider that uses an active screening process to view all files before they are uploaded to the site demonstrates that the service provider has apparent knowledge of infringing activity.[129] Here, the service provider’s employees not only reviewed each file uploaded to its site but they were also well aware of the popularity of the content and the fact that the content owner enforced its copyright.[130] Further, the fact that a bulletin board operator tracked or had the ability to track user uploads and downloads sufficiently demonstrated that the bulletin board operator had apparent knowledge of infringing activity taking place on its site.[131]
Also, apparent knowledge may be imputed where a service provider turns a “blind eye” to infringing activities taking place on its site. This is because “[w]illful blindness is knowledge, in copyright law.”[132] Thus, “[t]he knowledge element…is met … where a party has been notified of specific infringing uses of its technology and fails to act to prevent future such infringing uses, or willfully blinds itself to such infringing uses.”[133] For example, apparent knowledge was imputed to a service provider where the service provider failed to update its email address with the Copyright Office which meant content owner’s notifications of claimed infringement went unheeded.[134] Thus, while the apparent knowledge standard under the DMCA requires a service provider to be aware of “red flags” of obvious infringement, a service provider cannot claim immunity under the Act by blatantly ignoring or willfully blinding itself to “red flags” of obvious infringement.
E. Does YouTube have apparent knowledge of infringing activity taking place on its site?
The facts overwhelmingly suggest that YouTube is “aware of facts or circumstances from which infringing activity is apparent” on its site.[135] First, a “brief and casual viewing” of YouTube reveals the infringing nature of the site and, thus, “red flags” of obvious infringement.[136] Specifically, YouTube has both “Top Rated” and “Most Viewed” web pages.[137] The videos featured on these pages often include unauthorized copyrighted material.[138] In addition, YouTube’s employees scour the site for video clips to feature on YouTube’s home page.[139] As noted above, active screening of content before it is uploaded to a service provider’s site may impute constructive knowledge.[140]
The facts suggest YouTube has more than just general knowledge that its site “may be the subject of online copyright infringement.”[141] In particular, it is well known that YouTube’s popularity is due to the posting of the unauthorized NBC video clip, “Lazy Sunday.”[142] Also, academics and media experts estimate that 30 to 70 percent of the content uploaded onto YouTube is without the consent of content owners and, thus, infringing.[143] In addition, at the time Google acquired YouTube for $1.65 billion, it set aside $200 million in stock for potential copyright infringement lawsuits.[144] The fact a substantial amount of money was set aside to contend with copyright infringement lawsuits suggests both Google and YouTube have apparent (if not, arguably, actual) knowledge of infringing activity taking place on YouTube.
In addition, YouTube has negotiated and continues to negotiate licensing deals with content owners in order to absolve liability for past and future infringement on its site.[145] YouTube would not have a sense of urgency to negotiate these licensing deals and give up some of its revenue, unless it knew there was some viability to the actual and threatened copyright infringement lawsuits. In fact, currently there are four copyright infringement lawsuits pending against YouTube.[146] YouTube has received and continues to receive thousands of copyright infringement notifications each month. For example, prior to filing its lawsuit, Viacom sent YouTube over 100,000 copyright infringement notices.[147]
In addition, YouTube has set a ten minute maximum limit on the length of video clips posted to its site.[148] The purpose of the limit is to prevent users from uploading entire movies. However, users still manage to upload entire movies to the site because they simply break up the movie into ten minute clips.[149] YouTube has not always had a maximum video length policy. On the contrary, when the site first went live video clips uploaded to the site could be of unlimited length. However, YouTube changed its practice in order to bolster its copyright policy.[150] The fact YouTube now limits the length of video clips further suggests that the company has apparent knowledge of the infringing activity on its site.
The press coverage of YouTube, including articles about the pending lawsuit, also suggests that YouTube has apparent knowledge of the infringing activity taking place on its site. In its answer to Viacom’s complaint, YouTube admits that Google’s CEO, Eric Schmidt, “stated in a media interview that Google would offer anti-piracy technologies.”[151] Further, YouTube has agreed and is already using filtering technology on its site for the companies that have entered into license agreements with YouTube.[152] The fact YouTube has employed filtering technology to filter out the unauthorized content for some companies is another fact that suggests YouTube is aware that infringing activity is apparent on its site.
The above facts overwhelmingly imply that YouTube is aware of, but chooses to ignore, “red flags” of blatant copyright infringement on its site.[153] Since YouTube is aware of infringing activity taking place on its site, YouTube has the burden to remove or block access to the infringing material in order to claim the DMCA protections.
F. Section 512(c)(1)(A)(iii): remove or disable access to material
Section 512(c)(1)(A)(iii) provides that if a service provider obtains actual knowledge of infringing material or apparent knowledge of infringing activity on its site, the service provider must “act[] expeditiously to remove, or disable access to, the [infringing] material.[154] The statute does not define what constitutes to act expeditiously. However, Congress intended to allow some flexibility “[b]ecause the factual circumstances and technical parameters may vary from case to case;” thus, making it impossible “to identify a uniform time limit for expeditious action.”[155]
The requirement to remove or disable access to infringing material must be fulfilled by a service provider, “even if the copyright owner or its agent does not notify [the service provider] of a claimed infringement.”[156] Compliance with section 512 (c)(1)(A)(iii) is critical for service providers because if a service provider does not adequately remove or block access to the infringing material on its site, then the service provider may lose the safe harbor.[157] As noted earlier, the importance of the safe harbor is that it bars the recovery of monetary relief and limits injunctive relief in the event a service provider is liable for copyright infringement.
VI. THE DMCA AND THE BURDEN TO MONITOR A WEBSITE FOR INFRINGING MATERIAL
Generally, the DMCA does not impose an obligation on a service provider “to seek out copyright infringement” on its network or system.[158] In fact, “a service provider need not monitor its service or affirmatively seek facts indicating infringing activity,” to claim any one of the safe harbor provisions.[159] But, Congress explicitly recognized that Section 512 “is not intended to discourage the service provider from monitoring its service for infringing material.”[160] Therefore, it is not unreasonable for a service provider to be required to monitor where the service provider has apparent knowledge of pervasive infringing activity taking place on its site.[161]
Section 512(c)(1)(A)(iii) implies that a service provider must monitor or police its site to qualify for the safe harbor provision. Specifically, the statute provides that a service provider with apparent knowledge of infringing activity taking place on its site must expeditiously remove or block access to infringing material, not infringing activity, to qualify for the safe harbor. However, at the time the service provider acquires constructive knowledge of infringing activity, the service provider is only “aware of facts or circumstances” from which infringing activity is apparent. This means that the service provider may not have knowledge of specific infringing material on its site. But, to qualify for the safe harbor, the service provider must remove infringing material. As a result, the only way a service provider with apparent knowledge of infringing activity on its site is able to remove or block access to specific infringing material is through monitoring or policing the site.
For example, it is plausible for YouTube to claim it does not have knowledge of specific infringing material on its site because YouTube may not know each time a video is uploaded to its site whether that specific video is infringing. That said, YouTube, as discussed above, is aware of facts or circumstances from which infringing activity is apparent on its site. The facts or circumstances include: (1) Google set aside $200 million when it purchased YouTube for copyright infringement lawsuits, (2) companies have asked YouTube to install filtering technology to deter and remove infringing content, (3) YouTube has several outstanding copyright infringement lawsuits, and (4) there are numerous newspaper articles and other press regarding the infringing activity taking place on its site. However, the facts or circumstances do not provide YouTube with knowledge of specific infringing material on its site, only apparent knowledge of infringing activity. But, the statute requires YouTube, even where YouTube only has apparent knowledge of infringing activity, to remove infringing material from its site. Thus, YouTube must monitor its site to locate and remove infringing material to qualify for the safe harbor.
The fact a service provider with apparent knowledge of infringing activity may have to monitor its site for infringing material is contemplated by the Act. Immunity under the DMCA is “granted only to ‘innocent’ internet service providers who can show they do not have a defined level of knowledge.”[162] Specifically, “[t]he DMCA’s protection of an innocent service provider disappears at the moment the service provider loses its innocence, i.e., at the moment it becomes aware that a third party is using its system to infringe.”[163] Thus, a service provider who is aware of infringement on its site is required to take action in order to remain within the safe harbor.
In addition, section 512(g) provides immunity to service providers from being sued by end users as a result of the service provider taking down infringing material. [164] The immunity in section 512(g) applies whether the service provider is taking down or blocking access to material due to a notification it received from a copyright owner or even where the service provider is taking down or blocking access to infringing material “based on facts or circumstances from which infringing activity is apparent.”[165] As a result, the immunity provision under section 512(g) provides an additional incentive for service providers to monitor or police their sites for infringing material in order to remove or block access to the material.
Congress did not craft the DMCA to provide service providers with a shield to free-ride off the content of others.[166] Thus, it follows that to remain eligible for the DMCA safe harbor protections a service provider should have to monitor its website for infringing activity where the service provider has apparent knowledge of infringing activity taking place on its site. Further, as one court has noted, “[t]he common element of [the DMCA] safe harbors is that the service provider must do what it can reasonably be asked to do to prevent the use of its service by ‘repeat infringers,’” which may include monitoring its site.[167] In other words, if a service provider can take simple measures (such as installing filtering technology) to deter the uploading of infringing material on its site or remove such material, then the service provider must do so.[168] As a result, a service provider that has apparent knowledge of infringing activity taking place on its site must monitor its site for infringing material to remain eligible for the safe harbor provision.[169]
The hesitation of service providers to monitor or police their sites is twofold. First, if a service provider monitors or polices its site, then its role is no longer a passive one and, as a result, it may sacrifice the DMCA safe harbor protection. Second, policing or monitoring a site may also impute apparent knowledge of infringing activity and, thus, unless the service provider ensures that all infringing content has been removed from the site, it may lose its DMCA safe harbor protection. But, Congress addressed this concern in the legislative history when it stated that “[c]ourts should not conclude that the service provider loses eligibility for limitations on liability under section 512 solely because it engage[s] in a monitoring program.”[170]
Further, as one court has noted, a service provider’s “voluntary practice of engaging in limited monitoring of its Web site cannot prohibit it from invoking the protections of the DMCA.”[171] Thus, for example, Revver.com voluntarily reviews every video clip made available for viewing on its site. [172] The fact that Revver’s monitoring program may impute actual or apparent knowledge to Revver does not preclude Revver from claiming the DMCA safe harbor protections. [173] But if Revver does not monitor or police its site and has actual or apparent knowledge of infringing activity on its site, then Revver will likely lose the DMCA safe harbor protections. As a result, a service provider with apparent knowledge of infringing activity on its site has a burden to police its site to locate and remove infringing material in order to remain eligible for the DMCA section 512(c) safe harbor provision.[174]
VII. YOUTUBE HAS THE BURDEN TO POLICE ITS WEBSITE FOR INFRINGING CONTENT
YouTube claims that it has no burden to monitor its site for infringing activity in order to enjoy the DMCA safe harbor protections. However, this is simply wrong. As noted above, once a service provider acquires apparent knowledge of infringing activity on its site, the service provider has a burden to monitor its website for infringing content and doing so will not result in the service provider losing its DMCA protections. The “facts and circumstances” show that YouTube is not an “innocent” service provider. Rather, YouTube is well aware of the infringing activity taking place on its site. Thus, YouTube must monitor its site and remove infringing content to remain eligible for the safe harbor.
The facts and circumstances overwhelmingly demonstrate that YouTube has apparent knowledge of the infringing activity taking place on its site. Specifically: (1)YouTube has received thousands of copyright infringement notices from content owners, (2)YouTube already uses filtering technology to filter out copyrighted content on behalf of specific content owners, (3)YouTube limits the length of its video clips to deter the uploading of entire movies, (4)YouTube has negotiated and continues to negotiate licensing deals to absolve itself of past and future copyright infringement liability, (5) at the time YouTube was acquired, $200 million was set aside for copyright infringement lawsuits, and, (6) currently, there are several outstanding lawsuits filed against YouTube for copyright infringement. Thus, it is unfathomable for YouTube to claim it does not have any apparent knowledge of infringing activity on its site.
YouTube is currently using filtering technology to filter unauthorized content on its site, but chooses only to do so for those content owners that have entered into licensing deals with YouTube. YouTube’s actions do not comport with the DMCA’s objective to foster cooperation between content owners and service providers. In addition, since YouTube is aware of the infringing activity taking place on its site, YouTube has the burden to filter out all infringing content on its site, not just the content for those companies that have entered into licensing deals with YouTube, if YouTube wants to claim immunity under the DMCA safe harbor provision.[175] As discussed above, the use of filtering technology by YouTube will not result in YouTube losing its eligibility for the section 512(c) safe harbor.[176]
VIII. CONCLUSION
The purpose of the DMCA is to provide “innocent” service providers with certainty and clarity as to their liability for copyright infringement. But, Congress did not intend that this purpose be achieved by sacrificing intellectual property rights. As Senator Leahy said, “we must make sure that our copyright laws protect the intellectual property rights of creative works available online” because “[o]therwise, owners of intellectual property will be unwilling to put their material online [and] [i]f there is no content worth reading online, the growth of [the Internet] will be stifled, and public accessibility will be retarded.”[177]
Congress crafted the DMCA to “preserve the strong incentives for service providers and copyright owners to cooperate to detect and deal with copyright infringements that take place in the digital networked environment.”[178] To facilitate cooperation, the DMCA requires those service providers with knowledge of infringement taking place on their sites – regardless of whether the service provider has acquired the knowledge from a notification of claimed infringement – to take action by removing or blocking access to the infringing material. In the case of those service providers with apparent knowledge of infringing activity, the service providers will have to monitor their sites in order to remove or block access to infringing material. As a result, the fact YouTube has refused to fully cooperate with copyright owners “to detect and deal with copyright infringements” taking place on its site by installing filtering technology is contradictory to the spirit and letter of the DMCA. Thus, YouTube should not be able to use the section 512(c) safe harbor as a shield to deflect copyright infringement liability.
Top of Page
Footnotes