When the Artificial Intelligence Law 2025 comes into effect, it will establish a modern governance mindset that focuses on management for development, placing people at the center while balancing innovation and risk control. The law is expected to usher in a new phase of progress, positioning AI as a sustainable growth driver for the country.
Promoting innovation, controlling risks
The Artificial Intelligence Law, passed by the 15th National Assembly on December 10, 2025, is Vietnam's first comprehensive framework law on artificial intelligence, comprising 8 chapters and 35 articles, effective from March 1, 2026. The law marks a significant turning point, making Vietnam one of the few countries with a specialized legal framework for AI, in line with international practices such as the EU AI Act – the world's first law on AI, effective from 2025; Simultaneously, it demonstrates a management orientation for development, balancing the promotion of innovation and risk control.
The Artificial Intelligence Law 2025 not only fills legal gaps but is also progressive, humane, and highly practical, clearly reflecting the viewpoint of the Party and State of Vietnam on putting people at the center in the digital age. These outstanding points deserve in-depth evaluation, as they not only protect citizens' rights but also create momentum for Vietnam to break through in the Fourth Industrial Revolution.
First, the law demonstrates a modern, flexible approach that aligns with global trends. The law applies risk-based management (instead of absolute prohibition), classifying AI systems into low, medium, and high risk, along with clearly prohibited behaviors. This approach is based on international legal frameworks but adapted to Vietnam, regulating only the output and usage behavior of AI, without deeply interfering with the internal algorithmic models of businesses. Thanks to this, the law limits obstacles to innovation, encourages startups, and promotes the freer development of technology businesses.
Another progressive point is the establishment of the National Artificial Intelligence Development Fund, mobilizing social resources to invest in computing infrastructure, shared data, and applied research. The law also prioritizes the development of a national AI strategy for each period, with annual implementation plans led by the Ministry of Science and Technology. In particular, based on this law, the Prime Minister will issue a list of essential datasets, prioritizing Vietnamese and ethnic minority languages, helping Vietnam build digital sovereignty and reduce dependence on external AI models. This is a strategic step that will help Vietnam not only "catch up" but also "surpass" in the global AI race, affirming its position as a nation with technological self-reliance.
Putting people at the center
The Artificial Intelligence Law 2025 stands out with its core principle of "putting people at the center." Article 4 clearly states: “AI serves humanity, it does not replace human authority and responsibility. It ensures the maintenance of human control and the ability to intervene in all decisions and actions of artificial intelligence systems” (Clause 2). This protects human rights and privacy, preventing “soulless” AI from dominating life.
The law emphasizes fairness, transparency, impartiality, and non-discrimination, while also requiring adherence to Vietnamese ethical standards and cultural values (Clause 3, Article 4). This is a deeply humane point, ensuring that AI does not harm society or spread foreign cultural biases.
The law also encourages the application of AI to serve social welfare: supporting people with disabilities, the poor, and ethnic minorities to narrow the digital divide; and preserving and promoting national cultural identity. Integrating AI education into the general education curriculum (computational thinking, digital skills, technological ethics) helps the younger generation approach technology responsibly and humanely.
Furthermore, the law clearly stipulates accountability by requiring AI providers and deployers to take responsibility for the consequences and to fully compensate for damages if incidents occur. These regulations not only safeguard individuals but also strengthen social trust in emerging technologies.
The law is designed to be concise, with a legal framework that is easy to implement without overlapping with other laws (inheriting from the Law on Digital Technology Industry, promulgated on June 14, 2025). Specific regulations, such as labeling AI-generated content (deepfakes, fake images, fake audio must have identifying marks), help users distinguish between real and fake, effectively combating fake news and fraud in practice. The requirement for transparency when interacting with AI (users must know they are talking to a machine) avoids confusion and protects consumer rights.
The law is also practical in its sandbox testing mechanism (controlled environment), partially exempting businesses testing sensitive AI from legal liability, helping to lower costs and encourage startups. For high-risk AI systems, only self-assessment and post-audit are required, without prior permission, reducing administrative procedures. At the same time, the law expands its scope to include foreign organizations operating in Vietnam, ensuring a level playing field and protecting digital sovereignty.
