Public bodies across the United Kingdom are struggling to cope with a surge in AI-generated complaints and requests that threatens to overwhelm already stretched services, according to reports emerging this week. The scale of the problem has grown so severe that government officials are privately warning of a crisis point approaching—one that could grind decision-making processes to a halt and cost taxpayers millions in wasted administration.
The Scale of the Crisis
The practice, which involves using artificial intelligence systems to automatically generate large volumes of complaints, Freedom of Information requests, and formal challenges, has exploded in popularity over the past eighteen months. What began as a niche tactic employed by activist groups has evolved into something approaching an industry, with commercial services now offering ready-made solutions for anyone wishing to flood public bodies with correspondence.
The implications are stark. While a single human can perhaps file a handful of legitimate complaints or requests in a day, an AI system can generate hundreds—sometimes thousands—of individual pieces of correspondence in the same timeframe, each technically compliant with formal requirements but collectively designed to exhaust administrative capacity. Public sector workers, already operating under severe budget constraints, find themselves spending precious hours processing material that may never have been written by a human being at all.
The financial cost to the public purse is difficult to calculate precisely, but estimates from affected bodies suggest each AI-generated request costs between £15 and £40 to process properly—money that could be spent on actual public services rather than feeding an algorithmic beast.
Who’s Behind the Flood?
The organisations deploying these tactics vary considerably. At one end of the spectrum sit campaign groups using AI to amplify legitimate grievances, arguing that resource constraints have historically prevented ordinary citizens from holding public bodies to account. At the other sit more concerning actors: commercial entities seeking to delay or complicate regulatory decisions, political operatives testing the limits of democratic processes, and potentially hostile actors looking to degrade public sector effectiveness.
Local authorities appear particularly vulnerable. With staffing levels still recovering from a decade of austerity cuts, many lack the infrastructure to distinguish between AI-generated correspondence and genuine human engagement. A Freedom of Information request generated by an algorithm looks identical to one written by a concerned resident—until you examine it closely, and by then the damage to service capacity has already been done.
The Legal Grey Area
Existing legislation governing complaints and FOI requests was never drafted with AI in mind. The Freedom of Information Act 2000, for instance, makes no distinction between requests submitted by humans and those generated programmatically—it cares only about whether the request relates to recorded information held by public authorities. This legislative gap has allowed the practice to flourish, technically within the letter of the law while arguably violating its spirit.
Privacy and data protection considerations present similar challenges. An AI system submitting a complaint is not acting as a data subject in any meaningful sense, yet it may be processing personal information about public sector workers, using information scraped from public records, or generating content that raises questions about authenticity and accountability.
What Officials Are Saying
Public response from officials has been characteristically cautious. Senior civil servants acknowledge the problem exists but are reluctant to be seen criticising any mechanism that might theoretically increase public engagement—even when that engagement is entirely artificial. One senior official at a major government department, speaking anonymously, described the situation as “deeply concerning but politically sensitive,” noting that any move to restrict automated requests would face accusations of suppressing accountability.
Local government representatives have been somewhat more forthright. The Local Government Association has raised concerns repeatedly about the impact on service delivery, with member authorities reporting that AI-generated correspondence now accounts for a significant and growing proportion of their administrative workload. The mismatch between service capacity and demand has become, in their view, unsustainable.
The Democratic Dilemma
Here lies the genuine paradox at the heart of this issue. Democracy depends upon citizens engaging with public institutions. Freedom of Information exists specifically to ensure that government remains accountable to the people it serves. How do you restrict AI-generated correspondence without simultaneously restricting legitimate public participation?
The question is not easily answered. Campaigners argue that wealthy individuals and corporations have always had advantages in navigating bureaucratic systems, and AI merely levels the playing field for those without dedicated legal teams. But this argument ignores the fundamental difference between a well-resourced human filing many requests and an algorithm generating thousands simultaneously—the latter doesn’t require resources, merely access to the right tools.
What Happens Next
The government faces pressure to act from multiple directions. Public sector unions point to the toll on workers who must process meaningless correspondence while genuine cases go under-resourced. Technology companies see a lucrative market in both generating and detecting AI content. And opposition politicians smell an issue that crosses traditional left-right divides—everyone, in principle, wants efficient public services.
Watch for potential legislative changes in the coming parliamentary session. Options on the table reportedly include mandatory disclosure when correspondence is AI-generated, financial penalties for systematic abuse of FOI processes, and enhanced powers for public bodies to reject requests they can demonstrate are part of coordinated automated campaigns. None of these solutions is perfect, and all will face legal challenges—but doing nothing is no longer a viable option.
The Bottom Line
Let me be straight about this: we’re watching the hollowing out of democratic engagement in real time. Not because people have stopped caring about their communities, their services, or their government—but because the tools we’ve built to amplify legitimate voice are now being weaponised against the very institutions they were designed to strengthen.
Public bodies aren’t perfect. They make mistakes, they withhold information they shouldn’t, they fail residents in countless ways. But drowning them in AI-generated pap doesn’t hold anyone to account—it just wastes resources that could be spent actually addressing the problems worth complaining about. The people most harmed by this aren’t the civil servants processing meaningless requests; it’s the vulnerable individuals whose genuine cases get lost in the noise.
Somewhere along the line, we decided that more engagement was always better engagement. We’ve been had. The algorithm doesn’t care about accountability. It doesn’t care about democracy. It cares about throughput. And until our political class wakes up to that distinction, the rest of us will be left cleaning up the mess—letter by meaningless letter.