More than 60 human rights and civil liberties organisations have called on the government to abandon plans to use AI to estimate the ages of asylum seekers. They warn the technology could wrongly classify vulnerable children as adults, with accuracy and discrimination risks heightened for older teenagers and ethnic minorities.
In an open letter to immigration minister Alex Norris, 62 groups - including Amnesty International, Human Rights Watch, Liberty, the Electronic Frontier Foundation, Foxglove and Open Rights Group - urged the Home Office to halt the planned rollout from 2027. The coalition argues that facial age estimation systems are too unreliable and potentially discriminatory to underpin asylum decisions.
The system is intended to help immigration officials assess whether people claiming to be under 18 are likely to be children or adults. Ministers have stressed that AI-generated estimates would be used alongside, rather than instead of, human judgement. Under current guidance, individuals are only treated as adults if two immigration officers independently conclude their appearance and behaviour strongly suggest they are significantly older than 18. Where uncertainty remains, they are treated as children and referred to local authorities. Campaigners say these safeguards fail to address fundamental flaws in the technology itself.
Accuracy concerns across ethnicity and age
Facial age estimation systems have shown varying accuracy depending on ethnicity and skin tone, the groups say, raising questions about their use for asylum seekers from minority backgrounds. Government documents acknowledge the technology is least precise for people aged 16 to 18 - exactly the group it would be used to assess. The margin of error could have significant consequences for young people seeking protection.
The coalition also points to the impact of trauma, conflict, malnutrition, sleep deprivation and hazardous journeys on physical appearance. "The faces of children seeking asylum tend to prematurely age from trauma and violence," the groups wrote, adding that such experiences may make them appear older than their actual age. They question whether the AI has been tested on individuals with those backgrounds and say there is insufficient evidence that it improves decision-making.
Transparency and unanswered questions
The Home Office has published only limited information about how the systems were developed and tested, the organisations said. They are seeking details on training datasets, accuracy evaluation methods, and the safeguards available if people want to challenge decisions. The groups have given the Home Office 21 days to answer questions covering testing procedures, training data, appeal rights and the role AI-generated estimates would play.
"We ask now that the Home Office take on this uphill task in good faith and release the information that is required to scrutinise the legality and ethics of this deployment of new technology on some of the most vulnerable people - and children - in our world," the groups said.
Why this matters for Government
For government professionals, this case highlights why deploying AI in sensitive areas like asylum processing demands careful policy oversight, not just technical readiness. Understanding the limitations of tools like facial age estimation is essential for those working in AI for Government roles. Resources such as the AI Learning Path for Policy Makers provide structured guidance on navigating the ethical and operational questions raised by such deployments.
Your membership also unlocks:
