UK Home Office May Face Investigation Over Use of OpenAI Tools
wired ·
A UK government watchdog is considering an investigation into the Home Office's use of OpenAI-powered generative AI tools to process asylum claims. The Independent Chief Inspector of Borders and Immigration has requested information from the Open Rights Group regarding these technologies amid concerns over their accuracy and secrecy. Researchers have found that error rates in the Asylum Case Summarization tool are significantly higher than official estimates, with no quality control for the policy search assistant. While the Home Office maintains that AI only assists trained caseworkers and does not make final decisions, critics argue that cultural biases and hallucinations could lead to discriminatory outcomes. Legal experts warn that failing to inform asylum seekers about the use of these tools may breach data rights laws. Consequently, both the Open Rights Group and the Public Law Project have submitted their findings to the Information Commissioner’s Office. These groups also highlight that several other high-risk AI systems are being used across the government without public disclosure.
wired