Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Copyleft for Alleviating AIGC Copyright Dilemma: What-if Analysis, Public Perception and Implications

Xinwei Guo, Yujun Li, Yafeng Peng, Xuetao Wei

TL;DR

This work investigates the AIGC copyright dilemma and proposes copyleft as a practical mitigation. Through a mixed-methods design—a formal what-if analysis, case studies, and a public perception survey—it demonstrates that no single ownership solution suffices and that a copyleft framework could alleviate conflicts while accelerating AIGC deployment. Quantitative results indicate substantial public openness to copyleft, especially with authorized use under loose restrictions, and highlight factors that influence willingness to adopt such licensing. The paper provides a concrete, stakeholder-inclusive pathway for governance, emphasizing staged responsibilities (data owners, model owners, model users) and mechanisms like watermarking, to advance AIGC-enabled productivity while mitigating copyright risks.

Abstract

As AIGC has impacted our society profoundly in the past years, ethical issues have received tremendous attention. The most urgent one is the AIGC copyright dilemma, which can immensely stifle the development of AIGC and greatly cost the entire society. Given the complexity of AIGC copyright governance and the fact that no perfect solution currently exists, previous work advocated copyleft on AI governance but without substantive analysis. In this paper, we take a step further to explore the feasibility of copyleft to alleviate the AIGC copyright dilemma. We conduct a mixed-methods study from two aspects: qualitatively, we use a formal what-if analysis to clarify the dilemma and provide case studies to show the feasibility of copyleft; quantitatively, we perform a carefully designed survey to find out how the public feels about copylefting AIGC. The key findings include: a) people generally perceive the dilemma, b) they prefer to use authorized AIGC under loose restriction, and c) they are positive to copyleft in AIGC and willing to use it in the future.

Copyleft for Alleviating AIGC Copyright Dilemma: What-if Analysis, Public Perception and Implications

TL;DR

This work investigates the AIGC copyright dilemma and proposes copyleft as a practical mitigation. Through a mixed-methods design—a formal what-if analysis, case studies, and a public perception survey—it demonstrates that no single ownership solution suffices and that a copyleft framework could alleviate conflicts while accelerating AIGC deployment. Quantitative results indicate substantial public openness to copyleft, especially with authorized use under loose restrictions, and highlight factors that influence willingness to adopt such licensing. The paper provides a concrete, stakeholder-inclusive pathway for governance, emphasizing staged responsibilities (data owners, model owners, model users) and mechanisms like watermarking, to advance AIGC-enabled productivity while mitigating copyright risks.

Abstract

As AIGC has impacted our society profoundly in the past years, ethical issues have received tremendous attention. The most urgent one is the AIGC copyright dilemma, which can immensely stifle the development of AIGC and greatly cost the entire society. Given the complexity of AIGC copyright governance and the fact that no perfect solution currently exists, previous work advocated copyleft on AI governance but without substantive analysis. In this paper, we take a step further to explore the feasibility of copyleft to alleviate the AIGC copyright dilemma. We conduct a mixed-methods study from two aspects: qualitatively, we use a formal what-if analysis to clarify the dilemma and provide case studies to show the feasibility of copyleft; quantitatively, we perform a carefully designed survey to find out how the public feels about copylefting AIGC. The key findings include: a) people generally perceive the dilemma, b) they prefer to use authorized AIGC under loose restriction, and c) they are positive to copyleft in AIGC and willing to use it in the future.
Paper Structure (19 sections, 11 figures, 2 tables)

This paper contains 19 sections, 11 figures, 2 tables.

Table of Contents

  1. Introduction
  2. Related Work
  3. The AIGC Copyright Dilemma: A What-if Analysis
  4. What if the data owner claims the copyright?
  5. What if the model owner claims the copyright?
  6. What if the model user claims the copyright?
  7. Summary of What-if Analysis
  8. Case Study Under the Copyleft
  9. Case 1: The New York Times sued OpenAI
  10. Case 2: Unsupported AIGC copyright claim
  11. Case 3: Supported AIGC copyright claim
  12. Public Perception: A Survey Method
  13. Results
  14. The AIGC Copyright Dilemma
  15. Copyleft is a feasible option
  16. ...and 4 more sections

Figures (11)

  • Figure : Figure 1: Familiarity with copyright across whether experienced AIGC. 1 = Never heard, 2 = Heard before, 3 = Limited knowledge, 4 = Familiar, and 5 = Very familiar.
  • Figure : Figure 4: Concern about AIGC copyright issue across different groups ("experienced AIGC: no" VS "experienced AIGC: yes" for the left panel, and "unfamiliar with copyright" VS "familiar with copyright" for the right panel). The x-axis has a scale identical to Figure \ref{['fig:2_c']}.
  • Figure : Figure 7: Level of agreement on using authorized AIGC and extent of freedom expected when using AIGC across different groups ("AIGC copyright infringement after copylefting: no" VS "AIGC copyright infringement after copylefting: yes").
  • Figure : Figure 1: Familiarity with copyright across whether experienced AIGC. 1 = Never heard, 2 = Heard before, 3 = Limited knowledge, 4 = Familiar, and 5 = Very familiar.
  • Figure : Figure 2: Copyright ownership of an AI-generated image across different groups (all participants, participants experienced AIGC, and participants familiar with copyright). a = van Gogh, b = da Vinci, c = Model user, d = Model owner, and e = Data owner.
  • ...and 6 more figures