Swedish Government proposes real-time AI facial recognition for police use (www.regeringen.se)

The Swedish government has proposed a new law enabling police to use real-time AI facial recognition technology to combat serious crimes such as violent shootings and explosions linked to criminal activities. This legislation, outlined in Prop. 2025/26:150, particularly addresses cases where immediate identification is crucial, including scenarios involving potential human trafficking victims, missing persons believed to be in danger, or suspects of serious crimes that carry significant prison sentences. The law also stipulates that the use of this technology must be proportionate and requires permission from a prosecutor or court, although emergency cases can start without prior approval, necessitating a subsequent request within 24 hours. This move is significant for the AI/ML community as it opens discussions on the ethical implementation of AI in law enforcement, stressing the balance between public safety and individual privacy. The introduction of real-time facial recognition raises critical questions about data security, potential biases in AI algorithms, and societal implications, given the robust capabilities of machine learning in image processing. As the law is set to take effect in July 2026, it highlights a growing trend of incorporating advanced technologies to tackle complex societal issues, while fostering a dialogue on governance and oversight in AI applications.