Forget tiny upgrades. Gen Z does not want a patched version of democracy or bureaucracy. They want a FORK—a government designed for their era.
In software development, “forking” refers to creating a divergent version of a codebase, taking the same foundation but developing it along fundamentally different paths for different users. This January, at the World Economic Forum Annual Meeting in Davos, Singapore effectively forked the state itself, unveiling a Model AI Governance Framework for Agentic AI. The framework offers a glimpse of what governance for Gen Z can look like: adaptive, technologically literate, treating uncertainty as a feature rather than a flaw, and structured for continuous iteration rather than frozen consensus.
The contrast with Nepal reveals incompatible operating systems for the state. In Nepal, older generations were forged through tangible struggle—the fall of monarchy, Jana Andolans, decades of economic precarity. For them, authority was embodied in party cadres and union halls; politics meant physical presence and face-to-face negotiation. Gen Z inhabits an entirely different world. They came of age on screens, their consciousness shaped by algorithmic feeds and transnational culture. Their politics emerge from comment sections before spilling into streets, their aspirations benchmarked against global standards in real time, their dissent articulated through memes as much as manifestos.
More fundamentally, Gen Z expects governance to operate like the digital platforms they grew up with: constantly updated, instantly responsive, transparently accountable, and designed around user experience rather than bureaucratic inertia. This shift mirrors what Jürgen Habermas described as a crisis of the lifeworld—a rupture in the unspoken assumptions that make mutual understanding possible. Lessons in political party discipline from older generations strike Gen Z as evidence of systemic paralysis, while their digital fluency reads to elders as performative or superficial. In effect, they are speaking entirely different political languages, shaped by fundamentally incompatible experiences of what politics is. Across Nepal, everyone wonders what Gen Z wants—but the answer will not be found at home. To see the future Gen Z wants, one must look to Singapore.
Singapore recently launched the “World’s First Guide for Responsible Deployment of Agentic AI.” This framework represents the fork, a deliberate redesign of governance for a generation that thinks in iterations, not monuments. The 27-page document addresses autonomous AI systems, agents that can plan, act, and adapt without constant human supervision. These systems don't just generate text; they take actions with real-world consequences. They access databases, execute transactions, and interact with other systems. When they malfunction, people get hurt. Money gets lost. Systems break.
What makes Singapore's approach a true fork is not technological sophistication. It is institutional philosophy. The document acknowledges uncertainty, admits that "best practices will evolve," and explicitly calls itself "a living document." This is the opposite of how Nepal's bureaucracy operates, where policies are carved in stone, rarely implemented, and revised only after catastrophic failure.
Consider accountability. Singapore recognizes that when AI agents act autonomously, traditional assignments of responsibility break down. Its solution is to establish chains of accountability upfront, define escalation procedures, maintain audit trails, and adapt as lessons emerge. Compare this to Nepal's approach: establish committees, draft terms of reference, and wait for consensus. By the time we have agreed on study group membership, the world has moved on. Singapore's framework was published within months. Nepal would still be debating the acronym.
Governance in Social transformations in Nepal
Singapore treats regulation as infrastructure, something to be built, maintained, and upgraded. The document integrates technical controls such as sandbox testing and monitoring protocols with organizational requirements like role definitions and escalation procedures. This kind of integration is alien to Nepal's bureaucracy, where technology is what the IT department handles and policy is what political appointees write.
This is not about AI governance. It is about state capacity. Singapore can publish a coherent framework for agentic AI because it has institutions capable of understanding complex technologies, consulting meaningfully with industry, and updating policies as knowledge evolves. The same capabilities underpin effective pandemic response, efficient public housing, and infrastructure that actually works.
This is what makes Singapore's framework a fork rather than advanced policy—it reveals a fundamentally different conception of what states should be capable of doing. The forked version treats the state as a learning institution, continuously upgrading its capacity. The original version, which Nepal still runs, treats the state as a fixed entity that occasionally updates when forced by crisis.
Nepal's bureaucracy will not be able to regulate AI agents because it cannot regulate much of anything effectively. The problem is not a lack of expertise—there are talented technocrats—but institutional dysfunction. Policies get drafted, approved, and ignored. The result is Governance Theater: impressive documents that change nothing.
Gen Z sees this clearly. They watch Singapore build digital infrastructure while Nepal's government struggles to maintain functional online portals. They observe Estonia offering e-residency while Nepal requires citizens to visit offices in person for documentation that could easily be handled electronically. They note that Taiwan's civic tech community shapes policy while Nepal's remains locked in analog processes. These are not random examples—they are different versions of the state, forked for different generations.
The economic consequences compound. Companies deciding where to deploy agentic AI will favor jurisdictions with clear regulatory frameworks. Singapore’s governance certainty becomes a competitive advantage. Nepal’s policy void becomes disqualification. We lose not only immediate investment but also the institutional learning that comes from regulating cutting-edge technologies.
Some will argue that Nepal faces more pressing challenges—poverty, infrastructure, basic services. This misses the point entirely. Effective AI governance requires the same institutional capabilities needed to address those fundamental challenges: assessing risk systematically, assigning responsibility clearly, implementing technical controls, monitoring outcomes continuously, and adapting based on evidence. Singapore’s framework exposes state capacity that Nepal desperately needs but lacks.
Singapore's framework demonstrates that effective 21st-century governance requires states to be learning institutions, capable of understanding complex technologies, consulting meaningfully, implementing adaptively, and updating as conditions change. Nepal's institutions possess none of these characteristics. We operate on pre-digital principles: hierarchical decision-making, static policies, implementation divorced from feedback, and change requiring consensus that never materializes.
The result is a legitimacy crisis Nepal's elite cannot comprehend. When young Nepalis compare Singapore's framework, published swiftly, technically competent, and institutionally sophisticated, with Nepal's inability to maintain basic government websites, they are judging whether their state is capable of navigating the future. The answer, increasingly, is no. Unlike previous generations, Gen Z can leave. They can work remotely for companies anywhere with functional digital infrastructure. They can participate in global labor markets that value competence over connection. Nepal's brain drain is not just about better salaries. It is about opting out of institutional dysfunction. They are choosing the forked version of governance over the legacy one.
This is the real meaning of forking the state. Singapore has created a version of governance that Gen Z can also recognize as legitimate, not because it is perfect, but because it operates on principles they understand: iteration over permanence, adaptation over consensus, transparency over mystification, and continuous improvement over static authority.
Meanwhile, Nepal's government remains focused on coalition management and constitutional disputes that matter to fewer people each year. We are not just falling behind technologically. We are becoming ungovernable to the generation that understands how governance should work.
The question is not whether Nepal will eventually adopt frameworks like Singapore's. Nepal will, probably decades later. The question is whether we will lose the talented young people we need to implement such frameworks, leaving us governed by those from the older generation who cannot imagine why any of this matters.
Singapore's AI governance framework is a mirror. What we see reflected is not a technological deficit. That can be remedied with investment and training. What we see is institutional incapacity so profound that we cannot even recognize what competent governance looks like anymore. More precisely, it requires recognizing that while we have been maintaining the legacy version of governance, updating incrementally, patching occasionally, and defending reflexively, other nations have forked the state itself, building versions optimized for different users with different expectations.
Nepal is never interested in making timely updates. They cannot because it would undermine the authority built on decades of institutional knowledge, knowledge that Gen Z finds utterly incomprehensible. But Gen Z has already made their choice. They see the fork. They understand the divergence. And increasingly, they are migrating to the other branch.
The tragedy is not a lack of resources or sophistication. The tragedy is our lack of institutional courage, the willingness to admit that governance itself must be forked, rebuilt from foundations compatible with a new generation, and optimized for radically different users. Until that admission is made, we will keep losing the very generation capable of building it. Without them, we are left running legacy code on obsolete hardware, bewildered by our own dysfunction while the world moves on.
Singapore is not just governing AI better than Nepal. It is forking the state itself, creating a version that Generation Z can recognize as legitimate. That is the framework Nepal has yet to adopt. Unfortunately, none of the political parties seem to care. They are trapped in the last century.
No one outside Gen Z, not even millennials, truly understands what this generation wants in Nepal. The proof is in Singapore. Singapore’s government leads the way, showing how to integrate agentic AI into governance that works for a cyborg generation.
(Shah is the author of the book ‘Algorithmocracy: Democracy in the Age of Bitcoin, Ethereum and ChatGPT’)
