June 6, 2017

Automated Copyright Enforcement Systems: Perils, Pitfalls, and Possibilities

Today, the Niskanen Center has released a paper examining the costs and benefits associated with a system of automated copyright enforcement. In recent years, software has been developed that makes automated identification of copyright infringements possible. Although these algorithms are less-than-perfect, their emergence in the digital ecosystem are kindling discussions of whether online service providers (OSP) ought to implement mandatory content filtration systems in order to qualify for intermediary liability protections under the Digital Millennium Copyright Act (DMCA).

Broadly speaking, these systems operate with content recognition software (CRS) at their core. Such software has been utilized by private OSPs as voluntary remediation mechanisms for addressing complaints over works flagged as potentially copyright-infringing.

Automated identification and remediation systems (AIRS) have been effectively utilized as mechanisms for addressing claims of potential copyright infringement, without necessitating lengthy legal battles. For instance, YouTube’s ContentID system affords rightsholders the opportunity to  monetize a share of the revenue of works. As the Niskanen Center’s adjunct fellow Regina Zernay and I note in the paper, the “voluntary adoption” of such systems “has largely helped ease many of the contentious issues that would otherwise have existed between rightsholders and content creators in the digital age.”

However, some have called for CRS to be mandatorily implemented, arguing that they can viably enforce copyright claims online. Such a system of automated copyright enforcement (ACES) could, they argue, help strike a more appropriate balance between the interests of rightsholders and content producers.

Although an ACES might, at some point in the future, hold the potential to serve a more important role in the enforcement of copyright claims, the current regime is actually working as intended. The compromise struck with the DMCA—between OSPs and rightsholders—has helped usher in an era of unprecedented content creation, and the ease with which content has proliferated online has been a boon to consumers and producers. Looking into the future, it may be possible that an ACES becomes a viable policy tool. In our paper, we examine the necessary technical hurdles to be surmounted for such a system to emerge, the pathways proponents may take towards implementation, and how such a system would impact how the court’s evaluation of copyright infringement claims may change as a result.

While the status quo may eventually unravel, we offer words of caution for embracing an ACES at this time. Eventually, such a system may be ideal, but for the time being, the old adage that “if it ain’t broke, don’t fix it” is an apt characterization of the current copyright enforcement system. We conclude:

In light of the effectiveness of the DMCA’s current safe harbor procedures, augmented by the proliferation of voluntarily-adopted AIRSs, for the time being, maintaining the current DMCA framework and promoting CRSs and voluntary solutions in support of this framework may be the most sound method of ensuring the continued progress of the useful arts.

From the executive summary:

[A]n [automated copyright management system] may help resolve two of the biggest shortcomings in the current copyright regime. First, it provides a more appropriate system of evaluation, as it addresses a technology-based problem with a technology-based solution. Second, it reshapes the existing private sector solution by closely applying the legal standard of infringement, providing a more accurate evaluation of what is (and is not) infringing. In short, it bridges the gap between the systems of creation and delivery, and the system of evaluation.

Notwithstanding the potential benefits of such a system, this paper also acknowledges that the technical complexities in creating such a system may make its implementation difficult. It also recognizes that, on net, the DMCA has been a benefit to both content creators and rightsholders, allowing for an explosive outgrowth of new content to dominate a newly interconnected world.

Ultimately, this paper concludes that an automated copyright enforcement system may eventually be a viable step in the evolution of copyright law. However, the creation of such a system may rest in the distant future, largely due to the technical limitations inherent in accurately detecting infringing content. Given the current ecosystem, we argue that promoting CRSs and private sector, voluntary solutions—all within the framework of the DMCA—is a far better alternative for ensuring the continued progress of the useful arts.

Read the full paper here.