TL;DR
Get everyday essentials delivered free with Prime
- Fast, free delivery on millions of items
- Prime Video, Amazon Music and more included
- Member-only deals all year
Sky News reports that written complaints and claims across several UK public services have risen since ChatGPT launched in November 2022. The figures show a broad pattern, but do not establish that AI caused the increases; officials and regulators say AI-written submissions are affecting public bodies’ work.
Sky News technology correspondent Rowland Manthorpe reported increases across nine channels, including a 144% rise in household complaints about water companies, a 78% rise in complaints about English local authorities and a 71% rise in written complaints to GP practices. The report also found increases in complaints to the Department for Work and Pensions, employment tribunals, legal services, the NHS and police forces. These figures compare volumes since ChatGPT’s launch; each channel may also have its own causes.
In a small test described in the report, an AI-text detector labelled 89 of 100 recent FOI requests as fully AI-written and nine as human-written. A smaller sample from 2019 was labelled entirely human. Manthorpe cautioned that the test was not conclusive: AI detectors can be unreliable, and the sample does not show that AI caused the wider rise in requests.
A survey of civil servants cited by Sky News found that 58% said public use of AI-generated content significantly affects their work. Respondents described receiving correspondence at speed and said some interactions felt like arguing with ChatGPT. Those accounts indicate reported effects on staff, while the survey figure does not by itself quantify how much time or money the extra work requires.
UK PUBLIC SERVICES · REPORTING SNAPSHOT
How AI Is Adding More Paperwork to Britain’s Bureaucracy
Written complaints, information requests and tribunal claims have risen across several channels since ChatGPT launched. The pattern is striking; the cause is not settled.
One reported rise, several channels
Change since ChatGPT launchSky News found increases across nine channels, spanning public services with different roles. The figures below show selected reported changes; each channel may have its own reasons for rising.
Other channels cited include the Department for Work and Pensions, employment tribunals, legal services, the NHS and police forces. No percentage is supplied here for those categories. Bars compare the three stated increases on a shared scale; the arrow indicates reported rises without a figure.
What the detector test found
A small test of 100 recent Freedom of Information requests was labelled by an AI-text detector. The result is a snapshot of the tool’s classifications, not proof of authorship.
Recent FOI requests · 100 sampled
2019 comparison sample
The report says detector results are inconclusive: tools can misclassify text, and this sample cannot explain the wider increase in requests.
Pressure on public service inboxes
Staff accounts · reported effectsMore letters can mean more work
Public bodies assess written submissions, clarify requests and prepare replies. Lower-cost drafting may add to that workload and affect response times, but the report does not quantify hours, delays or costs across services.
“It’s like being bombarded with correspondence at speed.”
Civil servants surveyed in research cited by Sky News“It honestly feels in some instances we’re arguing with ChatGPT.”
Civil servant surveyed in research cited by Sky News“It’s exhausting and can be quite anxiety-inducing.”
Civil servant surveyed in research cited by Sky NewsA similar rise across borders
Other examples add context to the discussion of automated submissions. They do not show that every complaint or claim was AI-generated.
Housing Ombudsman
Complaints rose from about 2,600 in 2022 to just over 7,000 last year, according to the report.
Consumer Financial Protection Bureau
Complaints rose fivefold over the period cited in the report.
Possible “agentic flooding”
Researcher Chris Schmitz identified 84 possible cases, as reported by TechCrunch in September. Most cases his team found involved people making claims they were entitled to make.
How FOI rules apply
ICO guidance · May 2026The Information Commissioner’s Office issued guidance after public authorities reported higher volumes and more complex requests, including submissions that misquoted legislation or needed clarification.
AI-written requests remain valid
Authorities do not need to determine who or what wrote a request.
Assess what is being asked
Handle the request through the usual public information process.
Ask when meaning is unclear
Officials may seek clarification if a request is poorly formed or unclear.
Keep routes accessible
Manage growing volume without making it harder to challenge public decisions.
What the figures cannot prove
Timing can point to a question, but it cannot isolate a cause. The report leaves important measures and explanations unresolved.
How much AI contributed: the parallel timing is suggestive, but it does not establish how much of the rise came from AI.
Whether submissions are abusive: the source does not say most complaints or claims are false or improper. AI drafting does not remove a person’s right to make a request.
The workload cost: no common measure of staff hours, response delays or expense is provided across the services.
What happens next: service problems, public awareness, changing demand and easier drafting may all play a role; future volumes remain uncertain.
Key questions
What has risen since ChatGPT launched?
Sky News reported increases in several channels, including water company and council complaints, written complaints to GP practices, and employment tribunal claims.
Does this prove AI caused the increases?
No. The report describes a broad pattern and a small detector test, but neither establishes cause.
Are AI-written FOI requests valid?
Yes. ICO guidance says they remain valid. Authorities may ask for clarification when a request is unclear.
Are the increased complaints shown to be abusive?
No. The source does not say most submissions are abusive or false. It notes that many people are making claims they are entitled to make.
Pressure on Public Service Inboxes
Public bodies process requests and complaints through systems built to assess written submissions. If people can produce more detailed or numerous letters at low cost, officials may face more work sorting, clarifying and answering them. That can affect response times and staff workload, although the report does not quantify those effects across the services listed.
The rise also carries a risk of overreaction. Many complaints and claims are legitimate, and using AI to draft them does not remove a person’s right to make a request. The policy challenge is to handle increased volume while keeping routes to challenge public decisions accessible.
As an affiliate, we earn on qualifying purchases.
A Similar Rise Across Channels
The figures span services with different responsibilities, from water companies and councils to GP practices and employment tribunals. Sky News says the volumes had changed little for years in several channels before beginning to rise around late 2022. The repeated timing is suggestive, but it cannot separate AI’s contribution from other changes in demand, service quality or public awareness.
The report describes similar developments outside Britain. Researcher Chris Schmitz identified 84 possible cases of “agentic flooding” across 11 jurisdictions, as reported by TechCrunch in September. The term refers to large-scale use of automated systems to submit complaints or claims. 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 saw complaints rise fivefold over the period cited. These examples provide context, but do not show that every submission was AI-generated.
In May 2026, the Information Commissioner’s Office issued guidance after public authorities reported higher volumes and more complex FOI requests, including submissions that misquoted legislation or needed clarification. The ICO said AI-written requests remain valid and authorities need not determine who or what wrote them. They can seek clarification when a request is unclear.
““It’s like being bombarded with correspondence at speed.””
— Civil servants surveyed in research cited by Sky News
What the Figures Cannot Prove
The available figures show that volumes rose over the period, but they do not establish how much of the increase came from AI. The report’s detector test covered only 100 recent FOI requests, and detection tools can misclassify text. It also does not provide a common measure of staff hours, response delays or costs across all the services.
The reasons for rising demand may differ by channel. The source material notes that water companies have faced difficult years, for example. It remains unclear how much the trend reflects service problems, broader public behaviour, easier drafting with AI or other factors, and whether volumes will continue to rise.
The ICO’s May guidance sets out how public bodies should treat AI-written FOI requests: they remain valid, and officials may ask for clarification if a request is poorly formed. The next practical question is how authorities adapt their processes to handle more correspondence without making it harder for people to exercise their rights.
The Sky News report does not identify a government plan or timetable for a wider response. Further evidence on request volumes, processing times and the reasons behind submissions would help show whether current guidance and staffing are sufficient.
Key Questions
What has risen since ChatGPT launched?
Sky News reported increases in several channels, including water company and council complaints, written complaints to GP practices, and individual employment tribunal claims. The reported changes are measured since November 2022.
Does the report prove AI caused the increases?
No. It describes a similar rise across several channels and a small detector test of FOI requests, but says the test is inconclusive and does not establish cause.
Are AI-written FOI requests valid in the UK?
According to ICO guidance published in May 2026, AI-written FOI requests are valid. Authorities do not need to determine who or what wrote them, though they can ask for clarification when a request is unclear.
Are the increased complaints shown to be abusive or false?
No. The source material does not say that most submissions are abusive or false. Researcher Chris Schmitz said most cases his team found involved people making claims they were entitled to make.
Source: Thorsten Meyer AI
Halloween Picks
halloween
As an affiliate, we earn on qualifying purchases.
