in Paul Torremans, ed., Intellectual Property and Human Rights (4th edition, Kluwer Law Int’l, forthcoming 2020), 709-744
Artificial Intelligence (AI) is increasingly and pervasively affecting society. AI and algorithms play a relevant role in the Intellectual Property (IP) discourse as well. Tomorrow, algorithms will of course come as creators and innovators but today they are already here as enforcers. Content sanitization online will soon — and inevitably — become the sole domain of algorithmic enforcement tools. This process towards algorithmic enforcement online has been characterized by a few important recent developments. First, emerging regulatory framework, new legislative obligations and judicial trends would like to force online intermediaries to implement architectural changes leading to sanitization of allegedly infringing content — possibly legit freedom of expression — by design. Second, as a consequence, algorithmic enforcement brings about privatization of enforcement and delegation of public authority. Third, in online enforcement of intellectual property — as well as any other content sanitization — algorithms take decisions reflecting policy’s assumptions and interests that have very significant consequences to society at large, yet there is limited understanding of these processes. Finally, but critically for the focus of this investigation, Online Service Providers’ (OSPs) regulatory choices — occurring through algorithmic tools — can affect profoundly enjoyment of users’ fundamental rights, such as freedom of expression, freedom of information and right to privacy and data protection. This Chapter will map the complex conundrum triggered by algorithmic enforcement in the IP domain with special emphasis on copyright and trademark enforcement online. In doing so, this chapter, first, briefly explains what algorithmic enforcement online actually is. Second, this chapter will dig into voluntary, judicial and legislative measures that have been promoting the emergence of algorithmic enforcement online. Third, this chapter will assess tensions between algorithmic enforcement and human rights, including due process, freedom of information, freedom expression, privacy and freedom of business. In conclusion, this chapter will discuss solutions for protecting human rights form algorithmic enforcement negative externalities.
Available at SSRN.