The UK will scan asylum-seekers’ faces for age checks—despite knowing the tech is flawed
Tests of age-verification technology show the risks of life-altering errors.
The UK government is moving forward with AI-powered facial age estimation (FAE) to verify the age of asylum seekers at the border, despite internal evidence that the technology is unreliable and biased.
The short version
- AI scans faces to predict age for asylum seekers arriving at UK borders.
- Internal reports show significant bias, particularly against Sub-Saharan Africans.
- Errors can lead to children being stripped of protections and placed in adult detention.
- The Home Office disbanded a scientific committee that warned the tech was "hideously inaccurate."
What happened
The UK Home Office plans to deploy FAE technology to "crack down on fake claims" by adults posing as minors. However, a leaked internal report reveals that the "best" performing algorithms tested showed substantial deviations, especially for Sub-Saharan Africans—the largest group of migrants crossing the English Channel. For women in this group, age estimates were off by an average of 4.6 years. Furthermore, the system's accuracy drops significantly when using the low-quality photos typically taken during initial border encounters.
Why it matters
This deployment carries life-altering risks. If a child is incorrectly classified as an adult, they may lose critical legal protections and be held in adult-only facilities. Critics argue that vulnerable populations are being used as test subjects for experimental, biased technology that the government continues to pursue despite warnings from its own scientific advisors.
Summary by Nerd News Network. Read the full article at Ars Technica via the links above and below.