AI Embedded Across UK Home Office in Near-Total Secrecy, Report Claims

AI has been embedded across the UK Home Office in near-total secrecy, reinforcing what researchers call a "digital hostile environment" for immigration, according to a report published by Open Rights Group (ORG).

The report, titled Automated Discrimination: Unpacking the Digital Hostile Environment, documents the use of artificial intelligence across at least six immigration functions: detecting potential sham marriages, estimating the age of asylum seekers, scanning the contents of phones belonging to new arrivals, conducting border surveillance, summarising asylum interviews and country policy information, and prioritising cases through the Identify and Prioritise Immigration Cases (Ipic) tool.

Border surveillance, according to the research, now relies on live facial recognition alongside AI-powered drones and satellites. The Ipic tool, which the report describes as opaque, is used to prioritise and assess immigration cases. AI systems also summarise country policy information used as evidence in asylum interviews.

The stakes are considerable. Immigration decisions determine whether people are granted protection or returned to danger, and the report argues that automating parts of that process — without public scrutiny — transfers risk onto some of the least powerful people in the system.

What does the report actually allege?

Open Rights Group assembled its findings from three limited sources: information the Home Office has published, freedom of information (FOI) requests, and parliamentary questions from MPs. The organisation states that AI is being trialled and tested in "near-total secrecy", with officials routinely invoking national security to refuse FOI requests and redact disclosures.

The report raises concerns about "reducing people to data points" and says AI is not neutral or accurate. It warns the technology can replicate bias and discrimination already present in society, particularly when a key driver for deployment is anti-immigration policy and rhetoric.

Successive governments come in for criticism. The report accuses them of pushing an ideology that reduces complex human cases to a technical problem that can be fixed.

"Relying on AI to decide who gets to stay here and who gets thrown back into danger is a deeply alarming move," the report states.

The phrase "hostile environment" carries specific political weight. The report links the current deployment of AI to the hostile environment policy first envisaged by Theresa May as home secretary in 2012 — a set of measures designed to make life in the UK difficult for people without immigration status.

Who supports the findings?

The report has cross-party backing from three parliamentarians:

• Bell Ribeiro-Addy, Labour MP
• Pete Wishart, Scottish National Party MP
• Siân Berry, Green Party

Ribeiro-Addy said: "In spite of questions and requests, the Home Office has been reluctant to publish the documents that would explain how AI is used in asylum processing, what the problems are and how they are being addressed. The department has not explained how it will ensure there is accountability when they fail or justified why some of the most vulnerable people, with the least power, should bear the greatest risks."

Sara Alsherif, a programme manager at Open Rights Group, said: "Migration is a human phenomenon and experience, not a technical problem that can be fixed by artificial intelligence. We urgently need transparency, accountability and strict regulation to restrict how AI is used and to prevent its wider rollout."

What is Open Rights Group demanding?

ORG is calling on the government to regulate AI and ban uses it considers undemocratic or risky for human rights. One specific target: the use of ChatGPT tools in asylum applications, which the group wants prohibited.

The organisation's wider demands include:

• More transparency around government AI systems
• Stronger freedom of information laws
• A narrower legal definition of national security
• A robust public register of government AI tools

The demand for a narrower national security definition goes to the heart of the report's evidence problem. ORG states that officials routinely invoke national security to stonewall FOI requests, which the group says is how near-total secrecy around AI trials has been maintained.

How has the Home Office responded?

A Home Office spokesperson defended the department's use of the technology, stating: "Use of artificial intelligence is just one of the tools at our disposal to protect our borders and does not replace or automate the role of human decision-makers."

The spokesperson added: "Asylum decisions are made by trained caseworkers, with AI tools used only to assist in navigating interview records and policy information. They cannot be used in isolation and do not make decisions on claims."

On oversight, the spokesperson said: "We regularly monitor the performance of these tools alongside our established quality-assurance processes and have already committed to publishing updated evaluations on their rollout."

The response addresses one specific use case — assisting caseworkers with interview records and policy information — but does not directly address the other applications documented in the report, including phone scanning, age assessment, sham marriage detection, facial recognition at borders, or the Ipic prioritisation tool.

The Home Office commitment to publishing updated evaluations, if delivered, would partially answer one of ORG's transparency demands. Whether it satisfies the report's authors remains to be seen.

Why the story matters beyond the UK

The dispute sits at the intersection of two unresolved policy questions: how governments should govern AI in high-stakes administrative decisions, and how much secrecy national security claims can justify. The Home Office case is a concrete test of both. As governments worldwide deploy AI in immigration and border systems, the UK's balance between operational secrecy and public accountability is likely to serve as a reference point — and the ORG report, backed by MPs from three parties, ensures the pressure for disclosure will not ease.