AI Overwhelms Government Offices | Complaints Surge Across 11 Countries: 2.7x in UK, 5x in US

What you'll learn from this article:

• A new problem called "agentic flooding" is emerging, where AI submits massive volumes of applications and complaints to government offices
• Complaints have surged 2.7x in the UK and 5x in the US
• Legitimate applications and spam are mixed together, and government agencies are struggling to keep up
• Japan is likely to face the same problem eventually

What Is "Agentic Flooding," Where AI Agents Overwhelm Government Offices?

Government services around the world are now facing a new problem: "agentic flooding." This refers to the phenomenon where AI agents (AI that performs tasks on behalf of humans) flood government offices with massive numbers of applications and complaints, overwhelming their processing capacity.

The term was coined by Chris Schmitz, a doctoral student at the Centre for Digital Governance in Berlin. According to his research, 84 cases have already been confirmed across 11 countries — meaning this problem is beginning to emerge worldwide.

Why is this a problem? Because government staff are running out of time to process applications, which could delay responses to people who genuinely need help.

Complaints Surge 2.7x in the UK and 5x in the US

Let's look at the specific numbers. The UK's Housing Ombudsman (a public body that handles housing-related complaints) saw complaints rise from 2,600 in 2022 to over 7,000 in 2025 — a 2.7x increase in just two to three years since the launch of ChatGPT.

The situation is even more serious in the US. Complaints filed with the CFPB (Consumer Financial Protection Bureau), which handles financial disputes, jumped from 3.2 million in 2024 to 6.6 million in 2025 — a fivefold increase.

Similar surges have been observed in filings at Brazilian courts and petitions to the German parliament. In other words, this is not a problem confined to any particular country — it is a global phenomenon.

Why Has This Happened? AI Makes Writing Easy

The reason is clear. The emergence of LLMs (AI that can write like a human) such as ChatGPT has made it easy for anyone to draft application documents and complaint letters to government agencies.

Until now, submitting a complaint to a government body required navigating complex procedures and understanding specialized terminology. As a result, many people gave up on filing even when they genuinely needed help. But with AI, anyone can simply input their situation and generate a proper application in minutes.

The UK's Housing Ombudsman has been receiving highly detailed complaints that cite specific clauses and reference complaint-handling regulations — clearly written by AI. Some of these even come from "no-win, no-fee" law firms that pre-screened cases using AI.

In short, the biggest driver of this surge is that AI has lowered the barrier to applying.

Legitimate Application or Spam? It's Hard to Tell

This is where a difficult problem arises. Among AI-generated applications, some are legitimate submissions from people who genuinely need help, while others are malicious spam.

According to the US CFPB, there are cases where credit repair companies (businesses claiming to improve credit scores) are using AI to generate mass dispute filings in order to clog up the system. This is clearly malicious.

On the other hand, there are also cases where people who previously had no recourse have, for the first time, been able to speak up with the help of AI — for example, immigrants with limited English proficiency or elderly individuals unfamiliar with legal terminology who can now assert their legitimate rights thanks to AI.

The UK's Housing Ombudsman has explicitly stated: "Applications should be judged on their content, and must not be rejected solely on the grounds that they were written by AI." Automatically treating AI-written applications as spam could itself become problematic as an instance of improper handling.

How Are Governments Responding? Strengthening Identity Verification

Government agencies have begun taking countermeasures. The US CFPB undertook a major overhaul of its systems in 2026.

Under the new rules, the following measures have been introduced:

• Mandatory two-factor authentication: In addition to a password, a confirmation code sent to a mobile phone is now required
• Identity and address verification: Checks to confirm that the applicant is a real person
• Prioritizing standard dispute procedures: Applicants are required to use regular channels first, rather than going directly to the government agency
• Declarations of truthfulness: Applicants must personally certify that the content of their submission is accurate

These measures are expected to reduce malicious spam. However, they also carry the side effect of making procedures more burdensome for legitimate applicants — a difficult balance to strike.

What About the Impact on Japan? Not Yet Major, But Preparation Is Needed

So what about Japan? At present, there are no reports of "agentic flooding" becoming a significant problem for Japanese government agencies. However, it is widely believed that the same problem could eventually emerge.

In Japan, the Digital Agency has been advancing a large-scale demonstration involving approximately 180,000 government employees using AI, beginning in May 2026. Meanwhile, local governments such as Hamamatsu City and Tokushima Prefecture have established subsidy programs for AI adoption. In other words, there is active momentum on the government side to use AI.

On the other hand, cases of citizens using AI to submit applications or complaints to government agencies do not yet appear to be particularly common. However, AI tools such as ChatGPT are beginning to see widespread use in Japan as well. What has happened in the UK and the US is likely to occur in Japan too, with some time lag.

The Ministry of Internal Affairs and Communications and the Ministry of Economy, Trade and Industry are preparing to update their guidelines on accountability for AI agents. The question of who bears responsibility for damage caused by AI is important not only for businesses but also for government agencies. Going forward, Japanese government bodies will also need to consider preparing for "agentic flooding."

Summary

• "Agentic flooding," in which AI sends massive volumes of applications to government offices, has been confirmed across 11 countries worldwide
• Complaints have surged 2.7x in the UK and 5x in the US, exceeding the processing capacity of government agencies
• Legitimate applications and spam are mixed together, making them difficult to distinguish
• The US has introduced countermeasures such as strengthened identity verification, but these come with side effects
• Japan is likely to face the same problem eventually, making early preparation necessary

While AI makes it easier for citizens to access government services, it also creates a new challenge by straining the processing capacity of government agencies. How this problem is addressed looks set to shape the future of public administration.

---

This article is a cross-post from AI Friends.