39 The Digital Millennium Copyright Act
The rise of digital music recording technology in the 1980s, followed by the public internet in the 1990s, gave rise to a multitude of legal and technological challenges for the recording industry. New technologies created several new means for music consumers to make and distribute unlicensed copies of recordings, thus threatening the financial reward structure of the industry.
In a relatively rare and quick response to changing technology, the U.S. Congress passed a set of amendments to the U.S. Copyright law in 1998 designed to control the spread of digital piracy of music and video over the internet and to establish a legal framework for adjudicating the growing number of allegations of copyright infringement against internet providers. This set of amendments was named the Digital Millennium Copyright Act (DMCA) and has been an important source of both resolution and continued conflict as the internet has exceeded nearly all expectations of the extent to which it would transform the consumption and distribution of music and other creative content.
The most important part of the DMCA for our purposes is a set of provisions known as the “safe harbor” rules. These rules are contained in a set of new laws created by the DMCA and known separately as the Online Copyright Infringement Liability Limitation Act. (Other parts of the DMCA govern other issues related to copyright, computers, and the internet.) These safe-harbor provisions of the DMCA that relate to online infringement are contained in Section 512 of the U.S. copyright law, titled “Limitations on liability related to material online.”
The “safe harbor” rules provide a set of actions that an internet provider can take to shield themselves from copyright infringement claims, thus giving them a “safe harbor” from legal liability. Here are the steps the DMCA requires of an “internet service provider” (ISP) to take advantage of this legal shield:
- The ISP must not have actual knowledge that the copyrighted material or an activity using the material on the system or network is infringing;
- In the absence of such actual knowledge, the ISP must not be aware of facts or circumstances from which infringing activity is apparent; or upon obtaining such knowledge or awareness, acts expeditiously to remove, or disable access to, the copyrighted material;
- The ISP 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
- Upon notification of claimed infringement, the ISP responds expeditiously to remove, or disable access to, the material that is claimed to be infringing or to be the subject of infringing activity.
The term “internet service provider” is broadly defined to mean any person or company that either offers online access to a network or provides any services on such an online network.
The DMCA’s safe-harbor provisions constitute what is known as an “affirmative defense,” which means that a defendant in an infringement claim must affirmatively prove that they have met all the elements of the defense. In other words, it is not up to the plaintiff to prove that the defendant does not meet the safe-harbor threshold; rather, the defendant has to prove that they do meet that threshold.
Many companies rely on the safe-harbor provisions to allow users to provide online content of copyrighted material, most notably YouTube, but increasingly other services such as Amazon’s Twitch service, and TikTok. The safe-harbor provisions are not available to online companies that initiate the provision of copyrighted material themselves, such as Spotify. Those companies are not shielded from liability for any copyrighted material they place online without license to do so. However, for companies such as YouTube that rely on their users for placement of copyrighted material, rather than making it available themselves, the DMCA’s safe-harbor provisions are a crucial legal shield to copyright infringement claims.
YouTube and the DMCA
YouTube has become a high-profile laboratory for implementation of the DMCA’s safe-harbor rules. Let’s take a look at how YouTube stays in compliance with the DMCA’s legal safe harbor, which will in turn allow us to see where those rules may be creating new copyright problems while trying to solve the problem of online music piracy.
- User’s upload content and YouTube only provides the network. The most basic requirement for taking advantage of the DMCA safe-harbor provisions is to create a platform on which the content is created and shared by the user, rather than by the service provider (YouTube in this case). YouTube does not create video or music content (with the only very rare exception, such as YouTube’s infamous “best of YouTube” compilations); rather, it hosts a platform on which users post the content. YouTube plays the user content other users choose to view without altering that content.
However, we can also see in YouTube’s actions a potential flaw in their system: YouTube is not completely passive in its stance towards content. Rather, YouTube actively promotes and “suggests” certain content to users, and it “demonetizes” or outright bans certain content that it feels is outside the parameters of its guidelines. YouTube might claim that its guidelines are enforced and suggestions generated automatically through its “algorithms,” but those algorithms are not made public so suspicion arises that they are not as objective or automatic as YouTube claims them to be.
- YouTube removes access to videos when it becomes aware that they contain unlicensed copyrighted material. YouTube has put in place a well-known system by which the owner of copyrighted material can file a “takedown notice” that alerts YouTube of an allegedly infringing video. YouTube then offers the copyright owner a choice of whether it would like to have the offending video taken down or, in the alternative, “monetize” the video so that any ad revenue from the video gets redirected to their own account rather than to the user who uploaded it. By implementing this mechanism, YouTube retains its legal safe harbor under the DMCA by taking down infringing videos when it becomes aware of them (or allowing the copyright owner to claim any financial benefits from them).
There are two unfortunate consequences of this self-enforcement mechanism mandated by the DMCA. First, there is no mechanism to ensure that those who file these takedown notices are actually the copyright owners of the material in question, so many videos are falsely flagged for allegedly copyright violations. The takedown and monetization penalties are thus enforced with little to no fact-finding as to whether the claims are legitimate. YouTube does provide an appeal mechanism for users whose videos are issued takedown notices, but it’s unclear how many users are aware of their rights in such appeals, how long such appeals take, and whether the outcomes of such appeals accurately reflect the true ownership of the copyrighted material. There have also been cases of people filing fraudulent takedown notices against YouTube creators, falsely claiming that their videos are infringing copyrights, and then demanding payment from those users to withdraw their claims lest the users lose monetization of their content or face a ban from the site from multiple violations. In 2019, YouTube successfully sued one individual who had repeatedly filed false takedown notices in such an effort to extort payments from YouTube creators. The individual was forced to apologize for their fraud and was fined $25,000. (YouTube vs. Christopher Brady, USDC, Dist. of Nebraska, 2019.)
- By removing access to the infringing videos, and banning repeat offenders, YouTube does not financially benefit from copyright infringement on the site. YouTube’s takedown notice and content removal mechanism allows YouTube to comply with the DMCA’s requirement that a provider must not financially benefit from any copyright infringement on its site in order to obtain the legal safe harbor.
These issues were addressed in an important legal dispute between Viacom (and other plaintiffs) against YouTube initiated in 2007, in which Viacom claimed that YouTube infringed on the plaintiffs’ copyrighted video and audio content by hosting videos that contained that content. In 2010, the District Court held that YouTube was protected from legal liability under the DMCA safe-harbor rules, and that judgement was affirmed in part and reversed in part by the 2nd Circuit Court of Appeals in 2012. Before the District court could reconsider the issues reversed by the Court of Appeals, however, the lawsuit was settled by the parties. The 2nd Circuit Court of Appeals opinion issued in 2012 provides some insight into how YouTube manages to retain its legal safe-harbor despite the presence of unauthorized copyrighted material on its site.
The Viacom decision first affirms that the “actual knowledge” component of the DMCA requires that the internet provider have knowledge of specific, unlicensed copyrighted material on its service, not just knowledge that there is likely some unlicensed material somewhere on the site. Thus, YouTube (and other online providers) would only have to address the specific unlicensed copyrighted material that it becomes aware of, rather than having the duty to be sure that there could not theoretically be any unlicensed material on the site. YouTube’s “copyright takedown notice” system, which asks users to identify unlicensed material they find on the system, gives YouTube the actual knowledge of such specific infringements without having to concern themselves with potential infringement that might be present in the system. This application of the statute seems reasonable and practical: we would not want to require a service provider to hunt down potential copyright infringement among millions of videos because that would likely be impractical. Instead, it is more practical to only ask the provider to respond to actual instances of copyright infringement that it becomes aware of.
The DMCA and Fair Use.
The DMCA was enacted by Congress in an attempt to catch copyright law up to the internet age. In particular, the safe-harbor provisions of the DMCA discussed above provide an extra-judicial process and guidelines for internet service providers to deal with copyrighted content being posted by users on their services. Those guidelines have generally served to limit the amount of unlicensed copyright material appearing on the internet, thus protecting the rights of copyright holders. However, because this process skirts the judicial system and relies instead on the actions of internet service providers in a self-policing mechanism, some of the protections that have evolved to protect content creators from over-aggressive enforcement of copyright laws have been pushed aside.
In particular, the set of protections that come under the name “fair use” have taken a back seat to the efficiency of the “safe-harbor” protections. As we learned above, fair use is an argument that a defendant can raise in a legal proceeding when their content may technically violate a copyright, but when that violation is excused because the purpose or non-commercial nature of the content weighs in balance against its restriction due to a lack of a license. However, when anyone can file a copyright take-down notice on YouTube, and the result is an immediate takedown or demonetization of the content, the opportunity to assert a fair-use defense is often lost, or at least delayed or encumbered.
A copyright takedown notice system is not a legal proceeding: there is no judge, no jury, no witnesses, and no legal representation. The content creator may have some ability to appeal the takedown notice, but that appeal is also not heard in a judicial setting. Recall that a fair-use defense is a factual argument that is meant to be considered by a judge and jury after witness testimony and legal argumentation. Those process protections are only an after-thought in the DMCA safe-harbor guidelines, leading many content creators who might have very legitimate fair use arguments left to merely wonder why they have no forum in which to make those arguments in a system that is clearly tilted in favor of copyright holders.
The issue of fair use in the context of DMCA takedown notices was considered by the 9th Circuit Court of Appeals in a well-known 2016 decision, Lenz v. Universal Music, involving the music of Prince. The Lenz case was initiated by a mother who had posted a 29-second video to YouTube of her toddler son dancing to the Prince song “Let’s Go Crazy.” The focus of the video is clearly on the young boy and not on the particular music, which is coming from a phone or other low-fidelity audio source in the background. However, a Universal Music employee manually monitoring YouTube for copyright violations involving Prince’s music by searching for song names came across the video of the dancing toddler. The Universal employee, who promptly filed a takedown notice of the offending video, had not been instructed to consider fair use in making his determination, focusing instead only on whether a Prince song could be identified in the sound of the video.
Remarkably, the mother of the dancing toddler sued Universal for misrepresentation in its takedown notice, claiming that by not considering the potential fair use claims involved in the offending video Universal had not complied with the DMCA, which requires that a copyright holder have “a good faith belief that the use of the material in the manner complained of is not authorized by the copyright owner, its agent, or the law.” The 9th Circuit not only agreed with plaintiff Lenz, but clarified that because the fair use balancing test had been written into the 1976 Copyright Act — whereas previously it had only been a caselaw doctrine — it had thus been transformed from an affirmative defense into a right to use copyrighted material under certain circumstances. Given their determination that fair use is a right to use copyrighted material under certain circumstances, the 9th Circuit concluded that copyright holders must consider in good faith a potential fair use claim before sending a DMCA takedown notice.
However, despite the 9th Circuit’s Lenz decision, it is clear from the number of copyright takedowns of videos with readily apparent fair use arguments that copyright holders are not actually following the spirit of that decision. Part of that failure to consider potential fair use claims is certainly also due to the fact that DMCA claims have become mechanized under sophisticated “Content ID” algorithms that automatically flag videos for use of copyrighted material based on almost instantaneous recognition of a digital thumbprint of copyrighted sounds or images. Such automated systems, as sophisticated as they are, cannot yet account for potential fair use of the copyright material they find.
The issue of fair use rights being trammelled by the DMCA has begun to get a great of public attention in recent years, particularly as YouTube content creators have begun to complain that their use of copyrighted material for educational, criticism, and parody videos has been subject to takedown and demonetization by copyright holders who do not respect the fair use doctrine. Several of the more prominent music-education “YouTubers,” such as Rick Beato, Adam Neely, and Paul Davids have been vocal about their struggles to provide educational or critical commentary on popular music without facing demonetization or takedowns of their video content on YouTube, despite what they feel to be the obvious fair-use defense of their use of copyrighted music in those videos.
Criticism of the DMCA on several fronts, including its inability to adequately provide account for the fair-use defense, has increased to such an extent that the U.S. Senate Judiciary Committee, Subcommittee on Intellectual Property, has scheduled a series of hearings in 2020 to consider whether the DMCA should be revised. The issue of fair use was the topic of one such hearing on July 28, 2020, which heard from a range of witnesses, including copyright protection advocates and fair-use content creators (including YouTuber Rick Beato). One of the witnesses in these Senate hearings was Jane Ginsburg, professor of Literary and Artistic Property Law at Columbia Law School. In her testimony, Professor Ginsburg provided several possible fixes to the DMCA for Congress to consider in order to align the takedown procedures with fair use:
- Require copyright holders to conduct a fair-use determination before filing a takedown notice. In other words, prior to filing a takedown notice, copyright holders could be required to confirm that they had reviewed the allegedly infringing material for potential claims of fair use. Intriguingly, Professor Ginsburg alluded to the possibility that this could be done automatically through artificial intelligence, though she admitted it is unclear how that might work or whether that capability is currently feasible.
- Provide for an “alternative dispute mechanism” for content creators to appeal a copyright takedown based on evidence of fair use. Such a mechanism could be built into the process, allowing content creators to quickly access a mechanism to provide evidence of fair use, rather than having to first appeal a takedown after it has already occurred and wait for replies to their appeals that might still be in the hands only of the copyright holder.
- Encourage voluntary agreements between service providers and copyright holders that use of content below a prescribed minimal threshold percentage would automatically be deemed “fair use”. Ginsburg admits that such voluntary measures might be wishful thinking and less than accurate or predictable, but that they could be easily automated.
In his Senate testimony, YouTuber Rick Beato offered another solution that seems at least as capable of remedying the problem as those presented by Professor Ginsburg. Beato proposes that creators could apply to become certified “blue check” creators who have demonstrated that their content is made under a pre-cleared “fair use” intention and agreeing to follow certain guidelines to maintain that pre-clearance. Beato references Twitter’s similar “blue check” system for certifying its users to demonstrate the practicality of such a solution. The government already employs such pre-clearance systems in airline security with the TSA “precheck” flight security status.
Whether congress will eventually craft a legislative solution to the conflict between the DMCA takedown system and fair use is far from certain. Certainly, the current takedown situation is proving to be untenable for content creators who wish to create educational, critical, or parodic content on the internet using copyrighted material. Given the state of political paralysis currently gripping Congress, one would hesitate to bet on any quick resolution of any issue, but this issue does seem to be getting a great deal of attention in the public so anything is possible. Perhaps the level of attention to this issue will cause the internet service providers to craft a voluntary resolution of this issue in the hopes of avoiding legislative regulatory measures that they would find even more restrictive than what they can come up with themselves. At the time of this writing, it is far too early to tell which direction this will go.
