As artificial intelligence (AI) advances at a rapid pace, governments worldwide are racing to establish regulatory frameworks to oversee its development, deployment, and ethical implications. AI is shaping everything from healthcare and finance to national security and even democracy itself, making it more important than ever to balance innovation with responsibility. Governments worldwide are stepping up to regulate its impact, whether it’s the European Union’s AI Act, China’s strict algorithm rules, or India’s ethical AI guidelines. The reason is clear: without proper oversight, AI could lead to privacy breaches, security risks, and economic disruptions.
Nepal is no exception to this global shift towards AI governance. In early 2025, the country took a major step by introducing its first-ever draft AI policy, signaling its recognition of AI’s growing impact. While Nepal’s approach is still in its early stages, a comparative look at its framework alongside the EU’s AI Act reveals key differences in regulatory priorities, enforcement mechanisms, and the broader vision for AI governance.
Nepal Draft AI Policy: Development-Focused Regulation
Nepal’s AI policy takes a development-first approach, aiming to harness the power of AI to drive national progress. It envisions AI playing a key role in sectors like industry, commerce, information technology, transportation, defense, health, education, and agriculture. Rather than simply adopting AI innovations, this policy aspires to actively contribute to AI development, ensuring that its benefits are accessible, inclusive, and sustainable.
At its core, the policy is guided by a vision of building a more prosperous Nepal through the ethical and responsible use of AI. With a mission to leverage AI’s transformative potential for social and economic growth, it lays out several key strategies to achieve this goal:
Building a Strong Legal and Institutional Framework: Establishing a secure and sustainable AI ecosystem through appropriate legal and institutional frameworks.
Strengthening Infrastructure and Data Management: Developing essential infrastructure and efficient data systems to support AI research, innovation, and practical applications.
Encouraging AI Research, Innovation, and Entrepreneurship: Promoting AI studies, research, and entrepreneurship to drive technological advancement and position Nepal as a key player in the AI landscape.
Enhancing AI Production and Capacity: Supporting the sustainable development of AI technologies while ensuring long-term growth and efficiency.
Investing in Human Resource Development and Public Services: Promoting AI applications for human resource development, improving the quality of public services, and increasing productivity.
Expanding National and International Collaboration: Strengthening national and international cooperation to facilitate AI development and its secure application.
Ensuring Effective AI Regulation and Oversight: Establishing essential regulations and monitoring mechanisms to govern AI development and applications
To bring these initiatives to life, the policy proposes the creation of a Council, the National AI Center, and the AI Excellence Center, ensuring structured implementation and oversight.
One of the standout features of Nepal’s AI policy is its strong support for AI startups. By offering tax incentives and financial assistance, the policy fosters an innovation-friendly environment, helping local entrepreneurs thrive and contribute to AI-driven economic transformation.
At its core, Nepal’s AI policy takes a development-focused approach, striking a balance between innovation, regulation, and global cooperation. It aims to build a strong AI foundation that drives sustainable growth while ensuring that technology serves society ethically and responsibly.
EU AI Act: Risk-Based Compliance Model
Turning to the broader picture, the European Union’s AI Act stands out as a major milestone in the regulation of AI. It’s the first-ever comprehensive legal framework to adopt a risk-based compliance model, aiming to align AI systems with ethical standards, protect fundamental rights, and ensure public safety, all while encouraging innovation.
The EU AI Act categorizes AI systems into four distinct tiers, each with its own set of regulations:
Prohibited AI Systems: These are AI applications that infringe on fundamental rights, including social scoring, manipulative AI, and certain biometric surveillance.
High-Risk AI Systems (HRAIS): These are AI technologies used in sensitive sectors like law enforcement, hiring practices, and public services, which must comply with stringent rules to ensure safety and fairness.
General Purpose AI (GPAI): This category includes widely-used AI models like large language models. These must meet transparency and regulatory standards to ensure they’re developed responsibly.
Low-Risk AI Systems: These are the least risky AI applications, which require minimal regulatory oversight, mainly focusing on transparency.
The EU’s approach strikes a delicate balance, fostering innovation while ensuring accountability. By introducing this risk-based compliance model, the EU provides a clear regulatory framework that not only nurtures the growth of AI but also prioritizes safety and ethics. This model is particularly beneficial for small and medium enterprises (SMEs) and startups, ensuring that AI development happens transparently and responsibly.
Key differences and Recommendations:
While both Nepal’s draft AI policy and the EU’s AI Act share a common goal of advancing AI development with ethical governance, they differ in scope and regulatory priorities.
Risk-Based vs. Development-Oriented Approach
The EU takes a risk-based approach, focusing on regulating AI systems according to the potential risks they pose to safety, ethics, and fundamental rights.
In contrast, Nepal’s draft AI policy is more development-driven, prioritizing AI growth, innovation, and economic benefits while leaving room for regulatory flexibility as the sector evolves.
Data Security and Privacy Protection
The EU AI Act enforces strong data security and privacy protections. It requires AI systems to use privacy-compliant datasets and ensures transparency through strict audit mechanisms.
In contrast, Nepal’s Draft AI policy recognizes the importance of data security but currently lacks comprehensive legal safeguards to prevent potential misuse of AI, leaving room for improvement in protecting personal data.
Scope and Ambition
The EU’s AI Act has a broad reach, establishing strict, legally binding regulations, particularly for high-risk AI systems, with a focus on transparency and accountability. Meanwhile, Nepal’s draft policy is more geared toward positioning the country within the global AI ecosystem, fostering innovation, and promoting research, with less immediate regulatory pressure.
Focus on innovation
Nepal’s draft policy places a stronger emphasis on innovation. It aims to cultivate a supportive environment for AI startups and fuel growth through infrastructure development and international partnerships.
The EU, on the other hand, centers its approach on a regulated framework to ensure AI’s growth adheres to safety and ethical standards, with somewhat less focus on nurturing open innovation compared to Nepal’s proactive approach.
Way Forward: Making AI Work for Nepal
Nepal’s AI policy is a great step forward, but the real impact comes from action, not just plans. To make AI truly work for people, three key areas need focus.
Turning Vision into Reality
AI must be more than theory, it needs real-world use. Nepal should create local AI hubs where students, businesses, and communities can test and apply AI in areas like farming, healthcare, and public services. Practical solutions will drive real change.
Making AI Education Simple and Useful
AI shouldn’t be just for tech experts. Short, hands-on courses in local languages, AI boot camps, and community programs can help people use AI in daily life. Training must be simple, practical, and open to everyone.
Keeping AI Fair and Ethical
AI must work for people, not just profit. Nepal should set clear ethical guidelines so AI respects fairness, transparency, and human rights. AI startups in Nepal must integrate ethical AI principles from the ground up, ensuring their innovations are safe, unbiased, and aligned with both local values and global standards.
Conclusion
Nepal is on the right path with its AI policy, but policies alone won’t drive change, real action will. To truly benefit from AI, Nepal must focus on practical applications, ensuring AI reaches the people who need it most. This means supporting local AI hubs, making AI education widely accessible, and holding companies accountable for ethical AI development.
By putting people at the center of AI, Nepal can create a future where technology drives progress without leaving anyone behind. AI should be a tool for empowerment, innovation, and inclusivity, helping Nepal move forward while staying true to its unique cultural and economic landscape. The journey has begun and now it’s time to make AI work