“Vibe Governing” applies the idea of vibe coding to the public sector: when AI lowers the barrier to building, governments can focus on clearly defining problems rather than managing complex execution. With strong governance and oversight in place, they can move from stalled pilots to fast, accountable deployment—turning public intent into real solutions.
.png)
The rapid pace of artificial intelligence development makes it easy to forget how recently the term “vibe coding” emerged, and how quickly it reshaped expectations about who gets to build.
“There’s a new kind of coding…you let the vibes lead and forget that the code even exists,” wrote Andrej Karpathy on February 2, 2025, giving a name to what hundreds of developers had already experienced firsthand: the ability to rapidly build applications and products without writing the code itself.
This moment signaled that the glass ceiling had been shattered, a true technological emancipation that removed the primary barrier between idea and implementation.
This shift is most evident in a change in the dynamics of power, from those who know how to execute and build to those who can imagine, define, and precisely articulate what they want to create. In other words, those who truly understand a problem and can clearly frame and specify it are now far more likely to successfully build an effective solution.
This is an exciting democratization of capability. It works because it dramatically shortens the path required to produce a high-quality solution: those who define the problem can now also be the ones who create the solution themselves.
Among these newly empowered “problem solvers” stand governments and public agencies.
Their traditional mode of operation revolves around the axis of planning versus execution, with deep analysis of challenges, problems, and needs across the short, medium, and long term. As stewards of public resources, governments invest in extensive and complex planning and prioritization processes. One can debate how those priorities are sometimes set, but identifying problems and committing to address them are fundamental components of structured public policy.
Yet governments consistently struggle at the stage of execution and implementation.
They can be imagined as Gulliver, immensely powerful, yet bound hand and foot by thousands of small threads representing oversight, legislation, and accountability processes. These constraints are essential and meaningful, but they also render governmental responsiveness nearly irrelevant at today’s pace of change.
It is often assumed that governments are “not technological.” This is simply not true.
Governments and public agencies around the world are making extraordinary efforts to integrate innovative technologies into their services. They understand better than anyone that the end user, the citizen, expects services that meet the standards they are accustomed to receiving from their smartphones.
The growing number of AI pilots across local governments reflects a genuine desire to innovate. But it also highlights a systemic failure: agencies are often unable to move from pilot phases to real-world deployment and scale.
Recent studies point to exceptionally high failure rates for AI pilots in the public sector. Research from MIT found that approximately 95 percent of AI pilots fail to deliver their expected returns or value. A separate study by IDC presents a similar picture, showing that around 88 percent of proof-of-concept projects never progress to large-scale deployment, with only about 4 out of every 33 pilots reaching active production.
The reason is not technological, but structural: the absence of a controlled and secure production environment at organizational scale.
Without a robust governance layer that provides full visibility into AI usage, enforces policy, monitors risk, and ensures continuous regulatory compliance, any attempt to move fast becomes an institutional risk. In such a reality, governments cannot afford to operate in trial-and-error mode.
Viewed through the conceptual lens of vibe coding, it becomes possible to imagine what “Vibe Governing” could look like.
A government that deeply understands public needs would not merely manage problems, but actively create relevant, fast, and effective technological solutions on demand. When a need emerges in a specific community, or when an urgent public response is required, the government does not initiate a two-year procurement process. Instead, it distills the need and the intent, and the code is generated within an approved and governed environment.
Rather than investing energy in the “how,” meaning how to build a system, the government invests in the “what”: refining needs, listening closely to the public, and precisely defining desired outcomes. The government shifts from being a purely executing body to one focused on attentiveness and precision.
Examples of success demonstrate that this gap is not insurmountable.
In the United States, government departments that invested not only in AI models but also in infrastructure for governance, measurement, and enforcement have generated tangible public value. The New Jersey Department of Labor reported roughly a 35 percent increase in resident response rates through AI-enabled targeted communications. The state’s tax authority increased successful call resolution rates in service centers by approximately 50 percent.
The common denominator in these successes is not smarter AI, but clear frameworks of technological governance, frameworks in which organizational intent is translated into action within well-defined boundaries of accountability.
Still, governments cannot “vibe” in an open field where every mistake constitutes a national failure.
To truly free Gulliver, the threads that bind him must be replaced with an intelligent framework. This requires achieving technological sovereignty over the government’s own creation environment, a high-quality control and oversight layer that enables rapid action without undue risk.
It is the space where creative vision meets compliance and accountability, allowing governments to move beyond pilots into real deployment by providing the protected laboratory in which innovation can safely occur.
It is clear that we are witnessing a genuine transformation.
But as with any revolution, it is worth asking: whom does it serve?
Public institutions carry undeniable responsibilities that must be taken with the utmost seriousness. At the same time, they also bear responsibility for continuously improving how they serve the public.
The government that succeeds in harnessing the democratization of code as a driver of social creation will be the one that establishes a new, secure operating environment that sets expectations for its workforce not only to identify problems, but also to imagine and build the necessary solutions themselves.
This is Vibe Governing in its true sense. Not governance driven by intuition, but governance capable of translating public intent into fast, measurable, and accountable technological action.
.svg)
.png)
