Report alleges Home Office is using AI for immigration decision-making

A new report claims that artificial intelligence is being used across the UK’s immigration system to create a ‘digital hostile environment’ for migrants.

The report from Open Rights Group (ORG) claims that the idea of AI governing immigration decision-making is “no longer rhetoric”. “AI is being woven into crucial parts of the immigration machinery,” according to their research, “and is poised to shape all of it.”

The research has come from Freedom of Information requests alongside interviews and questions from MPs. The report has the support of MPs from Labour, the SNP, and the Green Party.

The report claims that AI tools are being trialled, or actively used, to estimate the age of asylum seekers, summarise interviews, and prioritise cases. They can be used to search mobile phones and assess potential sham marriages. The report also alleges that live facial recognition, AI-powered drones, satellites, and surveillance towers are all being implemented at UK borders. Artificial intelligence systems can also be used “pre-empt unknown threats” allowing for “predictive” immigration enforcement, according to the research.

The ORG raise concerns that what they call an “automated fortress” use systems that are known to exhibit bias, saying that they are “encoding discrimination into the infrastructure of border control”.

Writing in one of the forewords to the report, Labour MP Bell Ribeiro-Addy notes that: “People seeking safety are not a risk to be scored, a data point to be calculated, or a backlog to be managed.” She says that the report “exposes how quickly artificial intelligence is spreading through the UK's immigration and border system, and it asks questions that the government has too often avoided. We know the names of the companies that build these systems. We know far less about what they do.”

A spokesperson for the Home Office told The Guardian newspaper: “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. 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.”

Ribeiro-Addy notes, however: “Ministers insist that these AI tools are used to support decisions, rather than replacing human judgement. However, a summary does not need to be the source of a decision to influence it profoundly.”