Towards AI Copyright Equilibrium

Research output: Contribution to journalArticleScientificpeer-review

Abstract

To balance generative AI (GenAI) innovation with the protection of copyright for authors and performers, it is necessary to recalibrate the concept of "public interest." This recalibration is crucial to ensure that authors and performers receive fair and equitable remuneration for their contributions while facilitating public access to knowledge and cultural expressions. Such a redefinition is also aimed at addressing current challenges, including fair use, open access, and the democratization of information within the AI industry. Drawing on Virginia Held’s typology of public interest theory, this article proposes that adjustments to the notion of public interest should include: establishing a balance through either a majority of individual interests or empirical data; aligning with the collective interests that receive societal endorsement; and evaluating public interest based on normative content and moral judgment, utilizing the principle of enjoyment and the public perception test in copyright law. While various theoretical frameworks could be used to conceptualize public interest, the article proposes an approach that explicitly defines copyright objectives and harmonizes the rights of authors and performers with the public's right to access creative works. Such harmonization could be achieved through an integrative methodology that combines evidence-based analysis, consensus among stakeholders without conflicting interests, and normative evaluations rooted in societal ethics.
Original languageEnglish
Pages (from-to)1-17
JournalTalTech Journal of European Studies
Publication statusAccepted/In press - 5 Jan 2025
MoEC publication typeA1 Journal article-refereed

Keywords

  • Artificial Intelligence
  • Copyright
  • Innovation
  • Public Interest
  • Creativity
  • EU

Field of science

  • Law

Fingerprint

Dive into the research topics of 'Towards AI Copyright Equilibrium'. Together they form a unique fingerprint.

Citation for this output