AI Is Flooding Britain’s Bureaucracy With Paperwork. Friction Won’t Stop It.

🔍 Read the full analysis: AI Is Flooding Britain’s Bureaucracy With Paperwork. Friction Won’t Stop It. on ThorstenMeyerAI.com

A Sky News report published October 20 documents increases in written complaints, Freedom of Information requests and employment tribunal claims across Britain since ChatGPT launched in November 2022. Technology correspondent Rowland Manthorpe says the similar rise across public-service channels is consistent with AI making formal writing easier, though the data does not prove AI caused the increases.

Sky News reported increases across nine categories, including household complaints about water companies, up 144%, complaints about English local authorities, up 78%, and written complaints to GP practices, up 71%. It also reported increases of 64% in complaints to the Department for Work and Pensions, 62% in individual employment tribunal claims, and 35% in written NHS complaints. The figures compare levels since ChatGPT launched; the supplied report does not provide a single shared endpoint or detailed methodology for every category.

The pattern, according to Manthorpe, is a long period of relative stability followed by rises beginning around late 2022. The channels have separate possible explanations: Sky News notes that water companies, for example, have faced difficult years. The timing and similarity across services are suggestive, but do not by themselves establish a causal link to generative AI.

Manthorpe also tested 100 recent FOI requests using an AI-text detector. The tool labelled 89% fully AI-written and 9% human; a smaller 2019 sample was labelled entirely human. He cautioned that detectors are unreliable and that this test cannot demonstrate AI caused the wider increase. Separately, an academic survey cited by Sky News found 58% of surveyed civil servants said AI-generated material from the public significantly affected their work.

More Claims Test Public-Service Capacity

The reported rise matters because public services must process these submissions, whether they arrive as traditional letters or AI-assisted forms. Higher volumes can add to workloads and delay responses, while more complex or poorly framed requests may require staff to seek clarification. Civil servants quoted in the survey described the experience as fast-moving and tiring, but the report does not quantify any resulting delays or costs.

The policy choice is not simply how to reduce the number of messages. The DWP figures reported by Sky News show the proportion of complaints upheld has fallen from the 50% range to close to 40%. That is a lower success rate, but it still indicates that some complaints are accepted; with a larger overall volume, the absolute number upheld may also be higher. The report presents this as a reason to avoid assuming that increased volume is mostly abuse.

If AI helps people express concerns or pursue benefits and appeals, adding hurdles could affect people with legitimate claims as well as low-merit submissions. The challenge for government is to handle more claims without making access harder for those entitled to seek a remedy.

FOI Rules And Automated Submissions

The Information Commissioner’s Office issued guidance in May after public authorities reported more FOI requests and more complex submissions. According to the Sky News report, the ICO said AI-written requests remain valid; authorities do not have to establish whether a person or AI wrote them, and can seek clarification when a request is unclear or badly formed. That guidance addresses the validity of requests, not the staffing or system changes needed to handle higher volumes.

The issue is also appearing beyond Britain. Researcher Chris Schmitz documented 84 possible cases of “agentic flooding” across 11 jurisdictions, according to reporting by TechCrunch in September. The cases reportedly show increases after 2022, although that research describes possible cases rather than proving a common cause. The UK Housing Ombudsman’s complaints rose from about 2,600 in 2022 to just over 7,000 last year; the US Consumer Financial Protection Bureau recorded a fivefold increase, as cited in the source report.

The next technical shift could be from chatbots that help draft a message to AI agents that submit requests or appeals for a person. Manthorpe demonstrated an agent service from Meta in the report. Automated submissions could make routine processes easier to use, while creating a sharper test for systems that depend on staff reading individual written accounts.

““It’s like being bombarded with correspondence at speed.””

— Civil servant quoted in the academic survey cited by Sky News

AI’s Role Is Not Proven

The report shows a timing pattern and examples of AI-written submissions, but does not establish that AI caused the rise in complaints, requests or claims. Each category has its own potential drivers, and the detector test is limited by the reliability of AI-text detection and its small samples. The figures also do not show, across all channels, how many submissions were written with AI, how many were justified, or how much extra staff time they required.

It is also unclear how quickly public bodies will face submissions produced directly by agents rather than drafted by a person using a chatbot. The report raises the possibility of automated appeals for parking tickets or rail-delay compensation, but it does not give a forecast for how widespread those systems will become or their likely effects on service capacity.

Service Design Faces The Next Test

The immediate next step for public bodies is to apply existing rules to requests and complaints, including the ICO’s guidance on FOI submissions, while monitoring workload and the quality of incoming material. The report does not announce a new government policy or a timetable for reform.

Loosemore’s proposed direction is to replace essay-style processes with services that collect structured information instead of relying on lengthy written accounts. That would require redesigning application and complaint systems, rather than trying to identify or block every AI-assisted message. Whether departments will adopt that approach, and how they will protect access for people with valid claims while managing larger volumes, remains to be seen.

Key Questions

What did the Sky News report find?

It reported increases in several written public-service channels since ChatGPT launched in November 2022, including water-company complaints, local-authority complaints, FOI requests and employment tribunal claims. The reported figures do not prove AI caused the increases.

Does the report show that most new complaints are AI-generated?

No. A detector labelled 89% of a sample of 100 recent FOI requests fully AI-written, but the report notes that AI detectors are unreliable and says the test does not establish why overall volumes rose. It does not measure AI use across all complaint categories.

Are AI-written FOI requests valid?

According to guidance from the Information Commissioner’s Office cited by Sky News, AI-written FOI requests remain valid. Authorities do not need to determine who or what wrote them, and can ask for clarification when a request is unclear.

Are the increased claims mostly frivolous?

The figures cited do not show that. The source report says the DWP share of complaints upheld fell from the 50% range to close to 40%, while also noting that a larger total volume could mean more complaints are upheld in absolute numbers. Schmitz said most cases in his research involved people claiming something they were entitled to claim.

What changes could public services make?

Tom Loosemore has argued for processes based on structured data rather than long written essays. The report does not say that UK departments have adopted this proposal or set a timetable for doing so.

Source: ThorstenMeyerAI.com