Introduction
The intersection of artificial intelligence (AI) and copyright law is a complex and evolving area, with significant implications for creators, legal systems, and the broader cultural landscape. The current legal framework primarily emphasizes human authorship as a requirement for copyright protection, raising questions about the extent to which AI-generated content can be protected under existing laws. This discussion is further complicated by differing international perspectives and ongoing legal developments.
Description
The “human authorship” requirement for copyright protection remains unchanged [7], with current evidence suggesting that AI systems do not provide users with sufficient control over the outputs generated [7], despite the potential for detailed prompts to influence expressive elements [8]. The US Copyright Office has affirmed that existing copyright principles can apply to generative AI, concluding that outputs can be protected only when a human author contributes sufficient expressive elements [5]. This includes scenarios where human-authored works are perceptible in AI outputs or where a human makes creative modifications [5], but not merely providing prompts [5]. The report emphasizes that merely inputting prompts into generative AI does not grant copyright over the resulting output [4], as this does not demonstrate sufficient control over the creative process [4]. Providing prompts is likened to offering unprotectable ideas under US copyright law [4]. While copyright may be available for human-created works used as inputs [7], creative arrangements of material [3] [7], and modifications of outputs [7], outputs generated solely by AI remain ineligible for copyright protection.
A recent decision from the Beijing Internet Court recognized copyright protection for an AI-generated image [7], citing the author’s intellectual achievements through the selection and modification of prompts [7]. However, it remains uncertain how this aligns with US legal perspectives [7], particularly regarding the treatment of prompts and modifications [7]. The evolving legal landscape surrounding AI and copyright is marked by ongoing litigation across various jurisdictions [7], with anticipation for forthcoming clarifications on issues of AI infringement and fair use [7]. This complex legal journey is likely to involve significant litigation in the US judicial system [2], as copyrightability is assessed on a case-by-case basis [2], requiring creatives to demonstrate their contributions to the work rather than solely relying on AI outputs [2].
In early 2023 [5], the Copyright Office initiated a comprehensive exploration of the intersection between copyright and AI [5], which included issuing registration guidance for works with AI-generated content [5], hosting public sessions [5], and gathering extensive public input [5]. The Office has clarified that artists can register copyrights for AI-assisted works, provided that human creativity is evident in the final product [3]. The Register of Copyrights emphasized that the protection of such works hinges on the presence of human creativity [3], stating that modifications or creative arrangements made by a human to AI-generated outputs can qualify for copyright protection [3]. However, any registration must include a disclaimer regarding the AI-generated content [8], and the determination of whether human contributions to AI-generated works are adequate for authorship must be evaluated on a case-by-case basis [9]. The Office finds no justification for amending existing laws to offer more protection for AI-generated outputs [5]. The forthcoming Part 3 of the Office’s report will focus on the legal implications of training AI models on copyrighted works, including licensing and liability issues [5]. The report highlights concerns regarding the use of copyrighted human works to train AI systems without permission [3], a contentious issue for creative unions like the Writers Guild of America [3], which has sought protections against the unauthorized use of their members’ works in AI training [3]. Issues surrounding the inputs used to train AI models [2], particularly regarding the potential unauthorized use of artists’ content from the web [2], remain largely unaddressed [2], but the implications of these rulings are expected to be significant in the future [2].
The report clarifies that using generative AI for ideation or brainstorming does not inherently affect the copyrightability of a final creative work [4], distinguishing between using generative AI as a substitute for human creativity and using it to assist or enhance human expression [4]. If a user references generative AI output for inspiration without incorporating it directly [4], this should not impact their copyright claim in the final work [4]. A key consideration in determining copyrightability is whether generative AI merely assists in the creative process or makes expressive choices in the final work [4]. The inclusion of AI-generated elements in a larger human-authored work does not diminish the copyrightability of the overall work [4], provided other copyrightability requirements are met [4].
While the question of whether text prompts can be copyrighted remains open [6], the Office likens prompts to instructions that convey uncopyrightable ideas but acknowledges that particularly creative prompts may contain expressive elements [6]. However, this does not extend to copyright protection for the works produced from those prompts [6]. The Copyright Office plans to monitor technological advancements and legal interpretations to ensure that its guidelines remain relevant for creators utilizing AI tools, reflecting the acknowledgment of AI’s role in copyrightable works [1], which has been positively received by AI filmmakers and creatives [1], despite ongoing debates regarding AI authorship [1]. Extending copyright protection to materials created by generative AI could promote the creation of more works of authorship [2], benefiting public culture and knowledge [2]. However, there are concerns that an increase in AI-generated outputs may diminish incentives for human creators [2]. Recent government initiatives aim to expand funding for AI technology [2], potentially benefiting large tech companies [2], while individual creators may struggle to secure support [2].
Conclusion
The ongoing dialogue between AI technology and copyright law presents both challenges and opportunities. As legal systems grapple with the implications of AI-generated content, the emphasis remains on human creativity as the cornerstone of copyright protection. The evolving legal landscape will require careful consideration of international perspectives, technological advancements [1], and the balance between fostering innovation and protecting human creators. The outcomes of this discourse will significantly impact the future of creative industries and the broader cultural ecosystem.
References
[1] https://venturebeat.com/ai/u-s-copyright-office-says-ai-generated-content-can-be-copyrighted-if-a-human-contributes-to-or-edits-it/
[2] https://www.polygon.com/news/517029/ai-art-us-copyright-report-ruling-human-effort-is-required
[3] https://www.euronews.com/culture/2025/01/30/new-report-clarifies-us-copyright-rules-for-ai-created-and-human-modified-art
[4] https://www.lowenstein.com/news-insights/publications/client-alerts/us-copyright-office-issues-report-on-copyrightability-of-generative-ai-outputs-ai
[5] https://www.copyright.gov/newsnet/2025/1060.html
[6] https://www.theverge.com/news/602096/copyright-office-says-ai-prompting-doesnt-deserve-copyright-protection
[7] https://www.jdsupra.com/legalnews/two-takeaways-from-the-u-s-copyright-1714031/
[8] https://www.jdsupra.com/legalnews/copyright-office-report-further-9470818/
[9] https://news.bloomberglaw.com/ip-law/copyright-office-report-outlines-eligibility-terms-for-ai-works-6