Luật Sở hữu trí tuệ sửa đổi lấp khoảng trống pháp lý về A.I | VTV24

Key Concepts

• Intellectual Property (IP) Law: The legal framework governing ownership of creative works, now updated to address AI.
• AI-Generated Content: Works created using artificial intelligence tools.
• Human Authorship: The requirement that a work must involve direct human creative input to be eligible for copyright protection.
• Intermediary Platforms: Digital platforms (e.g., social media, video-sharing sites) that host user-generated content.
• Data Training: The process of using large datasets to train AI models.
• Prompt Engineering: The act of providing instructions to AI to generate specific outputs.

1. Legal Status of AI and Authorship

The revised Intellectual Property Law in Vietnam establishes that AI is not recognized as an author. Copyright ownership remains exclusively with humans.

• The "Human Element" Requirement: For a work to be protected, it must demonstrate direct human creative effort. Simply using a prompt (e.g., "create a song in the style of Beethoven") is insufficient for copyright protection.
• The Challenge of Attribution: A significant legal challenge lies in distinguishing between AI-generated content (not protected) and human-created content (protected). Future regulations are expected to clarify the percentage of human vs. AI contribution required for protection.

2. Real-World Case Studies

• "Théâtre D’opéra Spatial": A painting created by AI that won a fine arts competition in Colorado, USA. The US Copyright Office denied its copyright registration, citing a lack of direct human creative input. This case sparked global debate regarding AI "plagiarism" of human artistic styles.

3. Training Data and Commercial Use

The law allows the use of publicly available data for research, testing, and training AI systems.

• The "Commercial" Ambiguity: While the approach is currently open and flexible to foster innovation, experts argue that the law needs to clearly define "commercial" vs. "non-commercial" purposes.
• International Comparison:
  • European Union: Implements strict regulations where authors can "opt-out," prohibiting their works from being used as AI training data.
  • Alternative Model: Some jurisdictions suggest a "pay-to-use" model where AI developers must compensate original creators if the training data is used for commercial gain.

4. Responsibilities of Intermediary Platforms

Article 198B of the revised law significantly expands the responsibilities of digital platform owners:

• Proactive Prevention: Platforms must implement technical measures (e.g., AI-based filters) to prevent the upload of copyright-infringing content.
• Cooperation: Platforms are legally required to coordinate with rights holders and state authorities to remove infringing content upon request.
• Shift in Liability: Previously, responsibility fell primarily on the individual uploader; now, the platform owner shares legal accountability for the content hosted on their site.

5. Expert Perspectives and Recommendations

Lawyer Nguyen Thanh Ha emphasizes the following:

• AI as a Tool: AI should be viewed as a "powerful tool" rather than a replacement for human creativity.
• Transparency: Creators should label their works to disclose the use of AI.
• Business Caution: Businesses operating online can no longer be "carefree" regarding copyright. They must verify the origin of their content, especially when commercial interests are involved.
• Future Outlook: The legal landscape is expected to evolve with more specific guidelines by 2025 to address the "gray area" between AI automation and human artistic labor.

6. Synthesis and Conclusion

The revised IP Law represents a progressive step in balancing technological innovation with the protection of human creativity. While the law provides a framework for recognizing human-led AI-assisted works, it leaves significant room for interpretation regarding training data and the precise threshold of human contribution. The shift in liability toward digital platforms marks a major change in the digital ecosystem, forcing companies to invest in better content-filtering technologies. Ultimately, the law encourages creators to embrace AI as a tool while maintaining a clear distinction between machine-generated output and human-authored intellectual property.