Concerns among creators
Three years ago, Xuan Huong, a young technician in Ha Noi, began exploring AI-powered music composition tools. Initially driven by curiosity about the new technology, she soon discovered AI’s remarkable ability to assist in the creative process.
Using simple prompts, AI systems can arrange, orchestrate and even generate vocals in a variety of styles. To date, Huong has created hundreds of songs, many of which have been purchased by businesses for communications and brand promotion purposes.
“If a copyright dispute arises one day, will I be protected by law? I may have written the lyrics, but the music was partly generated with AI assistance. I genuinely do not know who owns the rights to that product,” Huong said.
Her concern is shared by many people currently using AI in creative activities. If AI directly participates in the creative process, who should be recognised as the author — the user, the AI developer or the owner of the technology platform?
Many legal experts regard this as one of the most significant challenges AI poses to existing legal systems.
According to Professor Hoang Van Kiem, former Chairman of the State Professorship Council for Information Technology and Chairman of the Scientific and Training Council of Saigon International University, current intellectual property laws are fundamentally built on the principle of protecting human creativity.
However, with AI developing rapidly, an increasing number of products and inventions are being created almost entirely through automated processes, with little or no direct human intervention.
“This raises a host of legal questions. Can AI-generated products be protected in the same way as human inventions? If so, who should own those rights?” Kiem asked.
According to him, the issue is also being examined by the World Intellectual Property Organisation, which is seeking appropriate legal solutions. Some countries, including the UK and Japan, have begun proposing amendments to intellectual property laws to adapt to AI development. However, no comprehensive model has yet gained international consensus.
Building a legal framework for new challenges
While AI presents significant opportunities for digital economic growth, it also creates an urgent need for effective regulatory frameworks.
According to Tran Le Hong, Deputy Director of the Intellectual Property Office of Viet Nam under the Ministry of Science and Technology, there is currently no international agreement on how ownership of AI-generated products should be determined.
Some argue that rights should belong to AI developers because they create the underlying tools. Others contend that users should hold the rights because they provide the prompts, input data and creative direction.
In reality, the issue is even more complex, as multiple stakeholders may contribute to a single AI-generated product, including algorithm designers, software developers, platform owners, providers of training data and end users.
“There are numerous overlapping interests surrounding AI-generated products. That is why countries have yet to reach a common position,” Hong said.
He noted that Viet Nam has adopted a cautious approach. Existing regulations neither explicitly accept nor reject copyright protection for AI-generated works. Instead, they continue to follow the fundamental principle that protected intellectual property must be linked to human creative activity.
For patent applications, applicants must demonstrate that AI served only as a supporting tool rather than the actual creator of the invention.
Nevertheless, AI has been included in the amended Intellectual Property Law as an important policy direction. This reflects Viet Nam’s intention to encourage AI development while gradually studying legal mechanisms to address emerging issues.
One major challenge concerns the data used to train AI systems. Hong explained that the amended law permits the use of data for research, testing and AI training purposes, provided that it does not unreasonably prejudice the legitimate rights and interests of rights holders. However, determining what constitutes “unreasonable prejudice” is far from straightforward.
Modern AI systems are trained on vast datasets containing text, images, audio and other forms of content. Identifying the origin of data, the scope of its use and its impact on individual rights holders remains highly challenging.
“If protection is too strict and absolute, AI development will be severely constrained. But if regulation is too lax, the risk of infringing the rights and interests of data owners will increase,” Hong said.
This explains why many countries continue to study the issue rather than introducing absolute rules governing the use of data for AI training.
Beyond copyright and industrial property rights, AI is generating new challenges related to personal data, personality rights and the use of individuals’ images and voices.
In recent years, numerous cases have emerged in which the images or voices of celebrities were copied to create AI-generated fake content, sparking considerable controversy.
According to Hong, personal images are currently protected under the Civil Code, which requires consent before another person’s image can be used. However, legal provisions governing voice rights have not kept pace with advances in AI technology.
As AI becomes capable of reproducing the voice of an artist or individual with near-perfect accuracy, questions surrounding legal protection, commercial exploitation and liability are becoming increasingly urgent.
Another issue attracting attention is AI’s ability to imitate the creative styles of famous artists, painters and designers. Hong noted that current intellectual property law protects authors and specific works, but not artistic styles themselves.
“A style is a method of expressing ideas rather than a copyright-protected subject matter. Therefore, AI’s use or imitation of a particular artistic style cannot currently be considered copyright infringement,” he said.
Nevertheless, experts agree that the rapid development of AI is creating new demands for legal systems. Beyond copyright and related rights, issues involving personality rights, personal data protection and the allocation of legal responsibility in the digital environment will require further research and refinement in the years ahead.
