For decades, the primary defense mechanism of the modern bureaucracy has not been the law, but the paperwork. The sheer friction of filing a formal complaint—the labyrinthine forms, the precise legal phrasing, and the daunting administrative hurdles—acted as a natural filter. Most citizens simply gave up before they began. But in the last twenty-four months, that filter has vanished. Government offices across the West are suddenly waking up to a deluge of perfectly formatted, legally sound, and highly persistent petitions that are overwhelming their existing capacities.
The Data Behind the Administrative Deluge
This sudden spike in activity is not a localized anomaly but a systemic shift across multiple continents. In the United Kingdom, the Housing Ombudsman has seen a dramatic escalation in the volume of grievances. In 2022, the agency handled approximately 2,600 complaints; by last year, that number surged to over 7,000. The trend is even more pronounced in the United States. The Consumer Financial Protection Bureau (CFPB) reported a fivefold increase in complaints over the same period, a jump that has strained the agency's ability to process claims in a timely manner.
Similar patterns are emerging in other democratic jurisdictions. Brazil has seen a spike in judicial petitions, and Germany is reporting a significant rise in parliamentary petitions. These are not isolated incidents of seasonal growth but a synchronized leap in civic engagement. Researcher Chris Schmitz has been tracking this phenomenon, which he terms agentic flooding. In a forthcoming paper to be presented at the AI Ethics and Society conference, Schmitz details his analysis of 84 potential flooding cases across 11 different jurisdictions. His findings indicate that while submission rates remained relatively stagnant prior to 2022, they began a steep, vertical climb coinciding exactly with the mass adoption of large language models (LLMs). The data suggests this growth is not a temporary bubble but a permanent shift in how citizens interact with the state.
From AI Spam to the Democratization of Rights
To understand why this is happening, one must distinguish this trend from the AI spam that has plagued the private sector. In the early days of the LLM boom, software companies were overwhelmed by a flood of low-quality bug reports and feature requests generated by bots. Those were empty shells—noise designed to mimic signal. Agentic flooding is the opposite. The surge in public agency filings consists largely of legitimate claims from actual citizens who possess valid legal rights but previously lacked the tools to exercise them.
The barrier was never a lack of grievance, but a lack of administrative literacy. For a low-income tenant in the UK or a defrauded consumer in the US, the cost of hiring a lawyer to draft a formal complaint was often higher than the value of the claim itself. Now, that barrier has collapsed. A citizen can simply take a photo of a confusing legal letter using the Claude app or feed a series of disjointed emails into ChatGPT to generate a professional, structured, and persuasive formal complaint in seconds. AI is acting as a pro bono administrative assistant for the masses, converting raw frustration into actionable legal documentation.
This creates a profound tension for the state. On one hand, the surge is a sign of a healthier democracy where citizens are finally able to hold institutions accountable. On the other hand, it is an operational nightmare. Public agencies are operating on budgets and staffing levels designed for the era of the paperwork filter. When the volume of legitimate claims increases fivefold without a corresponding increase in funding, the system reaches a breaking point. The challenge is no longer about filtering out the noise, but about scaling the capacity to handle the signal.
This shift forces a fundamental realization: the administrative burden was never a neutral part of governance; it was a silent gatekeeper. By removing that gate, AI has exposed the fragility of public service infrastructure. The current crisis is not caused by the AI itself, but by the gap between the new efficiency of the citizen and the legacy inefficiency of the state.
Public institutions must now move beyond treating this as a volume problem and begin redesigning the entire architecture of public service for an era where the cost of filing a claim is zero.
