Ethical AI has emerged as a pivotal global discourse, reflecting the growing realization of its profound impact on societies worldwide. While social issues give rise to policy solutions and thus legislations, ethics are deeply rooted in societal values that differ greatly from one jurisdiction to another. In the last decade, diverse groups and individuals have departed from using exclusively Free Software licenses on their projects to developing new types of licensing solutions which prioritize “ethical” restrictions on how software should be further used and distributed. In this talk, I will delve into the ethical dimensions of the convergence of AI and Free Software, examining the challenges and opportunities in building responsible and ethical AI solutions within the free software ecosystem.

This talk firstly provides examples of some of the most common licensing models imposing “ethical” restrictions. It then examines how licensing models imposing additional behavior restrictions that supposedly call themselves “open” are in contravention of the definitions of Free Software and Open Source Software. We will further explore some predominant material harms emanating from the use of these restrictive licenses. Lastly, given the various technical & legal obstacles in using these restrictive licenses, I implore that the ethical compliance checks must fall within the purview of regulations and not software licenses.

Given the proliferation of ethical licenses in AI landscapes causing additional behavior restrictions, integration of Free Software, license compatibility and licensing compliance become more complex, affecting the whole ecosystem. This presentation seeks to increase awareness about this crucial issue, particularly as public funds are being allocated to AI projects that may, at times, employ restrictive licensing practices.