THE UK Government has been accused of "hiding analysis" of AI facial recognition technology used to assess disputed ages of asylum seekers.
Following a freedom of information request from the SNP, Andy Burnham's government has refused to release background information on the software.
In May it was reported that a software company had been awarded a contract to develop and test the technology which will estimate a person's age by analysing photographs of them taken at the border. The contract will see the technology further tested and developed before being rolled out in mid-2027.
Human Rights Watch has urged the Government to scrap the scheme, describing it as "unproven technology" that will undermine the protections vulnerable children are entitled to.
The SNP asked the UK Government for information including data on the margin of error of the facial age estimation (FAE) technology during Home Office testing, any legal advice or policy assessments held by the Home Office regarding how technology complies with the European Convention on Human Rights, and any correspondence between the Home Office and COSLA or the Scottish Government regarding how FAE-generated age estimates will impact child placements transferred to Scottish councils.
But the UK Government has refused to provide any information, claiming it is exempt from disclosure as the policy is under development.
SNP MP Pete Wishart told The National that the public deserve answers on the risks of the technology, and urged Burnham to clarify whether he is happy to "ignore warnings from human rights and children's organisations".
He said: "Andy Burnham said we need to protect children from AI and this is a perfect example of how he can.
“These facial recognitions programmes rely on racially based algorithms that risk the safety of extremely vulnerable children.
“Real consideration needs to be given to if these systems are in any way ethical and Andy Burnham must be clear on whether he is happy to ignore warnings from humans rights and children’s organisations.
“It’s not good enough for the Home Office to hide analysis of the risks here from the public – it’s time we had answers.”
Burnham has previously called for tighter regulation of AI and has spoken about its risks for child protection.
He said in May: "How can a new wave of deregulation plausibly be the answer to the problems we have experienced since? This is … the kind of thinking that would doom us to repeat past mistakes and, if we’re not careful, prevent us from protecting children by failing to regulate social media, artificial intelligence and big tech.”
Angela Eagle said last year when she was border security minister that FAE "could produce workable results much quicker than other potential methods of scientific or technological age assessment" such as bone X-rays or MRI scans but "at a fraction of the cost".
The British Association of Social Workers (BASW) has warned that the government's plan to use AI in the assessment process will lead to major safeguarding mistakes.
Professor Sam Baron, interim chief executive of BASW, said earlier this year: "Assessing the ages of migrants is a complex process which social workers are best placed to do.
"This important task should not be open to shortcuts through artificial intelligence, especially as the pitfalls of getting it wrong can lead to major safeguarding risks."
Age assessments of asylum seekers are already carried out by border force officials who use methods such as examining documents, appearance, and demeanour to make an initial decision on age.
The new facial estimation technology will act as an additional tool to support officers at the border when a person's age is in doubt.
The Home Office has been approached for comment.
