Editorial illustration for UK Home Office May Face Investigation Over Use of OpenAI Tools
UK Home Office Under Investigation for OpenAI Use
The UK's Independent Chief Inspector of Borders and Immigration is weighing whether to open an inspection into how the Home Office uses generative AI tools built on OpenAI technology to process asylum claims, according to information shared with WIRED. The watchdog has reached out to the Open Rights Group for more detail on the tools as it finalizes which topics it will scrutinize next year, campaigner Sara Alsherif told WIRED.
That request follows a six-month investigation by Open Rights Group that included 11 parliamentary questions, two ministerial letters and 10 public records requests, all aimed at getting the Home Office to explain what these tools actually do and how well they work. The department has largely declined to share those details.
The scrutiny comes after lawmakers and researchers raised doubts about the accuracy of AI systems being used in a process that determines whether people get to stay in the country. The ICIBI flagged similar concerns last year, finding that one tool's performance was inconsistent. Whether the inspector adds this to its formal agenda remains undecided, but the request to ORG signals the question is already on the table.
A UK government watchdog is considering an investigation into the Home Office’s use of generative AI tools to process asylum applications, WIRED has learned. The potential probe follows lawmakers and researchers publicly questioning the tools’ accuracy and the government’s secrecy around their use.
Why this matters
This is a test case for what happens when government agencies bolt OpenAI's models onto decisions that affect people's safety and legal status, then treat the details as a secret. The ICIBI inquiry matters less for what it might find and more for the precedent: an oversight body actually pulling the thread on a deployment the Home Office would rather not explain. For builders shipping AI into regulated or high-stakes workflows, the lesson is blunt.
"We monitor performance alongside quality-assurance processes" is not a substitute for publishing what those processes actually measure, how often ACS and APS get asylum case facts wrong, or who checks the checkers. Open Rights Group didn't need a leak, it needed a watchdog willing to ask. If you're deploying language models to summarize or search case files where the stakes are deportation or asylum, assume someone outside your organization will eventually demand the error rates.
Better to have them ready than explained after the fact, under a headline you didn't choose.
Related Reading
Browse all OpenAI news →
AI Tools & Apps Claude gains shared context in Excel, PowerPoint; Microsoft adds Copilot Cowork
AI Tools & Apps Sarvam Signs 5-Year Defence AI Deal Covering Army, Navy and Air Force
AI Tools & Apps LG's recent webOS update adds Microsoft Copilot app, now removable
Related Topic OpenAI researcher quits, citing distrust over ad‑driven engagement metrics
LLMs & Generative AI
Related Topic OpenAI launches GPT-Image 1.5 with precise editing for enterprise visuals
LLMs & Generative AI
From Archives Canva’s Craft, first AI image model, plans designs before generating them
From Archives NVIDIA Minerva delivers dynamic, end-to-end observability for AI factories
From Archives OpenAI's Ted Sanders says model names can't be reduced to a single IQ score
LLMs & Generative AI
From Archives OpenAI turned coding tools into dots, an always-on personal assistant
Business & Startups
Common Questions Answered
What is the Independent Chief Inspector of Borders and Immigration investigating regarding the Home Office's use of OpenAI tools?
The ICIBI is weighing whether to open an inspection into how the UK Home Office uses generative AI tools built on OpenAI technology to process asylum claims. The watchdog has requested more details from the Open Rights Group as it determines which topics to scrutinize in the coming year.
Why are lawmakers and researchers questioning the accuracy of the Home Office's generative AI tools for asylum processing?
Lawmakers and researchers have publicly raised concerns about the accuracy of the AI tools used to process asylum applications, and they are also troubled by the government's secrecy surrounding how these tools are deployed. The Open Rights Group conducted a six-month investigation that prompted these questions about the tools' reliability and transparency.
What precedent could an ICIBI inquiry into the Home Office's AI use establish for other government agencies?
An ICIBI inquiry would set an important precedent by having an oversight body actually examine a government AI deployment that the agency would prefer to keep unexplained. This investigation matters as a test case for how oversight bodies can hold government agencies accountable when they use OpenAI's models for decisions affecting people's safety and legal status.
What lesson does the potential Home Office investigation send to organizations deploying AI in regulated or high-stakes workflows?
The investigation demonstrates that government agencies and organizations deploying AI in regulated or high-stakes workflows cannot rely on secrecy to shield their AI systems from scrutiny. The message is clear: oversight bodies may pull the thread on AI deployments, so organizations must be prepared to explain and justify their use of generative AI tools in sensitive decision-making processes.
Further Reading
- UK Home Office May Face Investigation Over Use of OpenAI Tools - WIRED
- AI tool used in thousands of asylum claims failing basic accuracy tests - The Independent
- UK secrecy over AI in immigration and asylum decisions under fire - Computer Weekly
- Open Rights Group raises concerns over Home Office AI asylum tools and calls for their suspension - Electronic Immigration Network
- Evaluation of AI trials in the asylum decision making process - UK Government
