AI-powered complaints are surging across the UK, overwhelming public services and raising concerns about the quality and legitimacy of grievances. Source: independent.co.uk.
AI Makes Complaints Easier — But More Problematic
Public bodies in the UK are receiving a sharp rise in complaints, many of which are now drafted using artificial intelligence tools like ChatGPT. Councils, schools, and police forces report a significant increase in the volume and complexity of complaints.
These complaints are often lengthy, filled with technical jargon, and contain inaccuracies. Before AI, writing a complaint required effort and clarity. Now, users can generate detailed grievances with a few prompts.
This ease of use has empowered more people to voice concerns — but at a cost. The sheer volume has created backlogs, while the quality of many has deteriorated. Experts warn this shift reflects a deeper issue in how citizens interact with public institutions.
How AI Is Changing the Complaint Process
AI tools generate coherent, well-structured text based on user input. When prompted with ‘complain about poor service,’ they produce a lengthy, formal document.
One example: a broadband complaint included a threat to Ofcom — a regulatory body — when it should have suggested an alternative dispute resolution process. This shows AI can misrepresent institutional roles.
In another case, the AI referenced an ombudsman that does not exist. This highlights a failure to verify factual accuracy.
Consumer advocate Helen Dewdney, who runs the blog The Complaining Cow, warns AI often misrepresents facts and fails to guide users to reliable sources. “First and foremost it gets a huge amount wrong,” she said. “I asked it to write an email complaining about broadband and lack of service. It didn’t advise on how to get the evidence, and in the email threatened Ofcom when it should be an alternative dispute resolution provider and how to check which one the company is a member of.”
She adds that AI is useful only when users provide accurate information and guide it properly. Without oversight, outputs can be misleading, legally unsound, or damaging to public trust.
Public Bodies Are Overwhelmed
Local councils are under growing pressure. Councillor Richard Clewer, chair of the Local Government Association’s public service reform and innovation committee, said AI-generated complaints are “longer and more complex” and are increasing administrative burdens.
He noted that local authorities are now considering potential government action to update regulatory frameworks to better manage the influx.
Case workers in MPs’ offices report that around half of the complaints they receive now contain AI-generated content. These are often verbose, lack key details like dates or incidents, and are difficult to assess. The lack of factual grounding makes it hard to tell if a complaint is genuine or frivolous.
Why This Matters
The issue is not just about volume — it’s about quality and fairness. Some complaints are genuine, from people who have been let down. Others are trivial or fabricated, simply to create noise.
This duality creates a paradox: AI empowers citizens to speak up, but also enables a flood of poorly constructed grievances that strain overburdened systems.
Journalist Sam Freedman described the situation as a “collective action problem.” He explained that for any individual, using AI to write a complaint is easier than before. But if every person does so, the system becomes overloaded, and no one gets a timely resolution. “For any individual, it’s easier to use AI. But if every individual does, everyone’s outcomes get worse because it takes so long to get a decision,” he said.
Delays are now severe — some cases are expected to be heard as late as 2029. This undermines the purpose of a complaint system: to deliver timely, fair, and actionable outcomes.
What Comes Next?
Experts warn public bodies may need to change how they handle complaints. This could include making it harder to file complaints or restricting AI use in response.
However, such changes could also reduce access for genuinely aggrieved citizens.
Freedman cautions: “Some AI-written complaints will be genuine so you can’t just filter out AI, you have to look at them. But then you start using AI to do that, and it becomes a situation where it is just chatbots talking to each other.”
There is no simple solution. The rise of AI in public complaints highlights a broader shift in how citizens engage with institutions — one that may strain the relationship between the public and the state. As AI tools become more accessible, public services must find ways to manage volume, verify content, and maintain trust.
Sources & further reading
Featured image: original publisher image. Via original publisher. Image source
