AI Agents Trigger 5x Surge in Public Service Requests as Agentic Flooding Spreads Globally

AI Agents Drive Unprecedented Surge in Public Service Applications

Public services worldwide are experiencing dramatic increases in applications and complaints as AI agents and tools like ChatGPT and Claude make it easier for citizens to navigate bureaucratic processes. Complaints to the UK housing ombudsman more than doubled from 2,600 in 2022 to just over 7,000 in 2025, while the US Consumer Financial Protection Bureau recorded 5x growth in complaints over the same period1

1

. Similar surges in government complaints appeared in Brazilian judicial petitions and German parliamentary petitions, signaling a fundamental shift in how citizens engage with public services.

Researcher Chris Schmitz is tracking this phenomenon as part of a broader trend called agentic flooding. In a paper co-authored with Lewis Hammond of the Cooperative AI Foundation and Alan Chan of GovAI, set to be presented at the AI Ethics and Society conference on October 12 to 14, Schmitz examined 84 cases of potential flooding across 11 jurisdictions2

2

. The research reveals broad evidence that AI-generated text is fundamentally changing citizen interactions with government agencies.

Understanding Agentic Flooding Through Quantitative and Qualitative Measures

The growth in agentic flooding manifests in two distinct forms: quantitative flooding, where the number of filings increases, and qualitative flooding, where individual submissions become longer and more detailed. Of the 84 cases studied, 50 showed quantitative flooding, 76 showed qualitative flooding, and 42 exhibited both patterns2

2

. One submission to a German social court exceeded 4,000 pages, demonstrating how language models enable citizens to generate extensive documentation that would have been prohibitively time-consuming before.

The researchers identified a clear pattern across most cases: submissions remained relatively flat before 2022, then rose at increasing speed as AI technology diffused throughout society. Crucially, most cases have not seen this growth slow down, suggesting the trend will likely continue rising for years to come1

1

. The study attributes 87% of the surges to a single underlying capability: language models can generate text cheaply and at scale2

2

.

Democratizing Access to Public Services or Creating New Challenges?

The research methodology was rigorous, starting with 2,288 government services across 12 countries and narrowing down to 84 cases that met three specific conditions. Each case required a plausible mechanism by which AI reduced the cost of applying, evidence that demand shifted, and an explicit statement from officials or credible third parties that AI caused the increase2

2

. In 58 of the 84 cases, government officials explicitly attributed the increase in filings to AI tools like ChatGPT and Claude.

Judicial and legal services accounted for 19 cases, followed by regulatory complaints with 10 and welfare and social security with 9. Services with payouts or benefits tied to successful claims, including tax returns, small claims, and property-value appeals, faced the highest near-term risk based on a 13-factor matrix developed by the researchers2

2

.

Legitimate Claims Versus Administrative Burdens

While the surge resembles what bug-bounty services experienced when flooded with low-quality AI-generated reports, Schmitz emphasizes a critical difference: most new applications to public services come from real people with legitimate claims. "The vast majority of cases we find are people who are entitled to claim for something, claiming for that thing," Schmitz told TechCrunch1

1

.

If these citizens weren't claiming benefits before, it may have been because administrative burdens made the application process too forbidding. Before, filing a complaint might have required dragging context together and prompting ChatGPT 3.5 very precisely. Now, it may be as simple as pasting or taking a photo of a letter with the Claude app and getting a quality response in one shot1

1

. This democratizing access to public services could ensure that entitled individuals actually receive the benefits they deserve.

Government Responses and the Risk of Excluding Vulnerable Populations

Governments are responding to agentic flooding through various measures. In 14 of the 84 cases, agencies responded by reinstating fees or blocking IP ranges, and Australia has considered reintroducing fees for Freedom of Information requests2

2

. Other potential responses include adding rate limits, in-person requirements, expanding capacity with more staff, deploying their own AI systems, or implementing redesigned intake processes.

However, the paper warns that measures like fees and rate limits can disproportionately affect poorer and less digitally literate applicants2

2

. Daily caps on submissions can limit high volumes but do little when a single applicant files one very large document. This creates a tension between managing resources and ensuring equitable access to public services.

The Opportunity to Redesign Public Services for the AI Era

Rather than viewing agentic flooding purely as a problem, Schmitz sees it as a rare opportunity to redesign public services for the AI era. "A big part of making AI go well is being able to detail out what the good version of things looks like. And anyone who's ever used ChatGPT to do the tax return knows that there's a good version here where you're being helped," Schmitz says. "This could be the moment to say, 'we need to rethink pretty much everything about how this process looks'"1

1

.

The researchers note that no AI agents currently exist that can autonomously navigate agency websites2

2

, suggesting that as technology advances, the impact could grow even more significant. The findings do not prove causation across government services as a whole and should be treated as proof that the phenomenon exists, not as a full measurement of its scale. Watch for how governments balance managing increased demand while ensuring that systemic reforms don't exclude the very populations that AI is helping to serve. The Bank of England has separately warned that autonomous AI systems could threaten financial stability, indicating broader concerns about AI's impact on critical infrastructure.