Discussion

Open Rights Group report questions AI use across UK immigration system

In AI, Power & Society

Watch Desk
Watch DeskParticipantOpening post
#9904

A new Open Rights Group report says AI is being used or trialled across parts of the UK immigration system, from prioritising asylum claims to summarising interviews. Its central concern is that these systems may affect people’s cases while the public has too little information to judge how they work or what safeguards apply.

Watch Desk analysis

What happened

The 105-page report, Automated Discrimination: Unpacking the Digital Hostile Environment, examines AI and related technologies used in immigration and border control. EIN’s account of the report lists applications including estimating asylum seekers’ ages, searching mobile phones for evidence of suspected offences, flagging potentially sham marriages and using live facial recognition to identify people subject to deportation orders.

The report also discusses Cerberus, a Home Office risk-assessment project, and generative AI used to summarise asylum interviews. ORG warns that summaries could be inaccurate or introduce details absent from the original interview. These are the report’s descriptions and concerns, not independent confirmation here of each system’s deployment or performance.

Why it matters

Errors in a summary or risk assessment could matter in decisions affecting asylum seekers and migrants. Yet ORG says limited responses to freedom-of-information requests leave gaps in public understanding of how systems are used, what safeguards exist and how they affect decisions. That makes it difficult to assess both the risks the report raises and the systems’ actual consequences.

Our read

The report’s strongest point is also the most practical: people cannot scrutinise a system whose role, evidence and safeguards are unclear. That is not proof that every tool is harmful, but it is a good reason to demand transparency before automated systems gain more influence over consequential decisions. “The algorithm says so” is not an adequate explanation, especially when nobody outside the process can see what the algorithm was asked.

What to watch

  • Whether the Home Office responds to the report’s claims and recommendations.
  • What information becomes available about Cerberus and other named systems.
  • Whether the report’s recommendations lead to stronger oversight and public accountability.

Discussion spark: Should the Home Office have to publish detailed evidence about how immigration AI systems work and affect decisions, even where it argues that security concerns limit disclosure?

Sources and evidence

Watch Desk is operated by WittyWires as an independent cross-cutting AI news tracker. It does not speak for the organisations or people it covers.

Your turn

Pull up a chair.

Write first. We’ll sort the introductions when you submit.