As technology advances at a rapid pace, the traditional boundary between human creativity and algorithmic output has become increasingly blurred. Generative Artificial Intelligence (AI) tools, capable of producing text, artwork, software code and music in seconds have sparked intense debate within the legal community worldwide. Central to this controversy is a fundamental question: How does modern copyright law adapt to an era where the creator may not be human?
Historically, legal frameworks governing copyright across international jurisdictions have operated on a core premise: protection is fundamentally reserved for human creators.
- The Originality Standard: To gain copyright protection, a work must display a minimum degree of human creativity, judgment and intellectual effort.
- The AI Dilemma: Outputs generated purely by autonomous algorithms struggle to satisfy the ‘human author’ requirement. Consequently, courts and copyright registries are routinely denying protection to fully machine-generated works.
When a user inputs a prompt into an AI tool and receives a refined creative output, determining who (if anyone) holds the legal copyright remains complex. Current legal discourse splits into three main interpretations:
- The User: The individual who conceived the prompt and provided creative direction through prompt engineering.
- The Developer: The organisation or software engineer that designed, trained and deployed the underlying model.
- The Public Domain: The position that autonomous generative outputs lack human authorship entirely and should immediately belong to the public domain.
Beyond the question of output ownership lies the critical legal hurdle of input data. Training modern AI models requires massive datasets, frequently scraped from the web and containing billions of copyrighted works. Key ongoing litigation revolves around whether using copyrighted media for machine learning constitutes non-infringing Fair Use or systematic, unauthorised exploitation.
To foster technological progress while continuing to safeguard the economic and moral rights of human creators, modern legal frameworks must proactively adapt. Potential policy and legislative solutions include:
- Sufficient Human Control Test: Establishing clearer legal benchmarks to evaluate how much human editing, arrangement, or creative guidance is required for an AI-assisted work to earn copyright protection.
- Statutory Licensing Models: Developing blanket licensing frameworks for AI training datasets to guarantee fair compensation for original creators.
- Transparency Directives: Enforcing mandatory disclosure and watermark regulations when legal documents, academic literature, or commercial media rely heavily on synthetic content.
The exponential rise of generative AI does not render copyright law obsolete; rather, it demands a thoughtful modernisation of existing principles. Legal practitioners, scholars and lawmakers must engage in robust, ongoing dialogue to build balanced legal frameworks that protect human ingenuity while encouraging technological innovation.
Written By: –
Sarasi Rivina Perera
(Faculty of Law)
(University of Colombo)
Designed By: –
Rtr. Munshifa Waseer
( Senior Blog Team Member 2026-27)
