‘AI clog’: UK public bodies battle onslaught of chatbot-generated complaints

Public bodies across the United Kingdom are now wrestling with a deluge of complaints, requests and legal challenges that have been drafted by generative‑AI tools. What began as a modest trickle in 2024 has swelled into what insiders are dubbing an “AI clog”, a phenomenon that is straining already thin resources in Parliament, local councils, schools and the courts. As senior caseworker Hollie Wickens of a Labour MP’s office puts it, roughly half of the emails her team receives are clearly AI‑generated, and the sheer volume and complexity of those messages are making routine work considerably more taxing.

The scale of the problem

Across the public sector, the surge in AI‑written correspondence is not a marginal inconvenience but a systemic pressure point. The Information Commissioner’s Office (ICO) reported a noticeable rise in the number of requests it received in 2025, noting that many of these were “very long, repetitive and contain unrelated comments or misdirected demands”. The length of a typical complaint has exploded: where a two‑page letter was once the norm, many now stretch to ten times that length, and the most extreme example cited by Lawyers in Local Government (LLG) reaches a staggering 200 pages.

For frontline staff, the impact is immediate. Hollie Wickens, who has a decade of experience in parliamentary casework, says the AI‑driven influx has turned what used to be a straightforward intake process into a labour‑intensive exercise in parsing, verification and response. The extra time required to untangle AI‑generated language – which often includes legal citations, bullet‑pointed arguments and verbose narratives – is diverting staff from other essential duties, compounding the chronic understaffing that plagues many public bodies.

What AI tools are doing

The rise of ChatGPT, Claude, Copilot, Gemini and similar platforms has effectively handed laypeople a free, albeit imperfect, legal adviser and administrative assistant. Users can produce a polished, formal‑sounding complaint in minutes, complete with statutory references and case‑law excerpts. As Deborah Evans, chief executive of Lawyers in Local Government, observes, “AI is a wonderful tool that enables people to do things very, very quickly and it can make a complaint sound incredibly professional, very formal.”

However, the speed and ease of generation come with a cost. The AI often pulls in jurisdictionally irrelevant material – for instance, citing Canadian case law when a UK council is the subject – and can hallucinate facts, inserting inaccurate or fabricated details. This creates a paradox for public servants: they must treat the correspondence as a legitimate request, yet spend additional time correcting errors and filtering out irrelevant content.

Impact on local authorities

Local councils are feeling the pressure acutely. Complaints now span the gamut from adult social‑care services to missed bin collections, and even routine planning objections. Councillor Alessandro Georgiou of Enfield warns that residents are increasingly using AI‑generated letters to appeal council decisions, noting a “significant uptick in parking appeals”. Yet, as Evans points out, councils receive no extra funding to cope with the surge; they must manage a larger workload with the same budget.

The sheer length of the letters is a tell‑tale sign for officials. The inclusion of dense bullet points, legal jargon and unusually long narratives often betrays an AI origin. While the content may be well‑structured, the time required to read, verify and respond to a 200‑page dossier is disproportionate to the underlying grievance, stretching staff capacity and delaying responses to genuine, non‑AI‑generated concerns.

The legal sector’s own flood

Lawyers in both the public and private sectors are reporting similar challenges. Employment tribunals, already grappling with a backlog, have seen a 28 % rise in claims opened between April and June compared with the previous year, amounting to roughly 14,000 new cases from single claimants. At the same time, the number of concluded cases fell, leaving about 70,000 open single claims by the end of June.

One technology start‑up founder, whose former employee is contesting a dismissal, illustrates the human cost. He predicts legal fees of around £30,000 for the tribunal case, a figure that underscores how AI‑enabled complainants can launch costly legal actions with minimal preparation. Jim Moore, an employee‑relations expert, notes that many AI‑generated complaints are “unnecessarily verbose” and tend to inflate minor incidents into major offences, often omitting the crucial question of what remedy the complainant seeks.

Why the AI “clog” matters

The phenomenon is more than an administrative nuisance; it threatens the efficacy of public service delivery. When staff are diverted to dissect AI‑laden letters, routine tasks such as processing benefits, handling genuine enquiries and maintaining essential services suffer. Moreover, the ICO cautions that even inaccurate AI‑drafted requests must be answered, meaning public bodies cannot simply dismiss them as spam.

From a governance perspective, the flood raises questions about the balance between technological empowerment and accountability. While AI promises productivity gains – a claim that has been linked to the stronger-than‑expected UK GDP growth this year – the unintended side‑effect is a new form of digital noise that can obscure legitimate citizen concerns and erode trust in public institutions.

Potential responses and the way forward

Stakeholders are beginning to explore mitigations. The concept of “AI slop” – low‑quality, high‑volume output that clogs systems – is now mainstream, prompting calls for better detection tools and clearer guidelines on AI‑generated correspondence. Researchers at the Hertie School’s Centre for Digital Governance, the Cooperative AI Foundation and GovAI have coined the term “agentic flooding” to describe the coordinated, automated inundation of public bodies.

In practice, organisations are training staff to spot tell‑tale signs of AI authorship, such as overly formal language, repetitive phrasing and the inclusion of irrelevant legal citations. Some are also considering policy measures, like requiring a disclaimer when AI is used to draft official complaints, though such approaches would need legislative backing to be enforceable. Until a robust framework emerges, the onus remains on overburdened public servants to navigate the AI‑generated deluge while maintaining the standards of service that citizens expect.

This article was produced with AI-assisted research and editorial support. Reporting is based on the source material cited below. Sources: The Guardian UK; theguardian.com; Global1.News (08 November 2026).

By Erica Thornton, Staff Writer