The Hague, 18 May 2020. UNICRI, through its Centre on Artificial Intelligence and Robotics, and INTERPOL's Innovation Centre in Singapore have released a new joint report "Towards Responsible AI Innovation" as part of their ongoing cooperation exploring the opportunities and potential pitfalls of artificial intelligence (AI) for law enforcement.
Building upon the foundations laid in their first joint report on the topic, this new report dives deeper into how AI can be utilized by law enforcement. The report analyses how AI can be used in four technical domains: i) resource optimization, ii) audio processing, iii) visual processing, and iv) natural language processing. The report further examines specific applications being explored by law enforcement in Australia, Germany, Japan and Norway. Complementing this, it also describes the concept of responsible AI in the context of law enforcement, focusing on lawfulness, social acceptance and ethics, and concludes with a recommendation for the preparation of a toolkit for responsible AI innovation to support law enforcement with the design, development and deployment of AI in a manner that is both lawful and trustworthy. The elaboration of such a toolkit will be the focus of the upcoming third INTERPOL-UNICRI Global Meeting on AI for Law Enforcement to be held in November 2020.
"Towards Responsible AI Innovation" can be downloaded here.
The new report was formally launched at a joint INTERPOL-UNICRI Virtual Discussion Room on 13 May that focused on the role AI can play in supporting law enforcement to preserve public safety and social order during the COVID-19 pandemic, as well as contain the spread of the virus.
In light of the changing nature of crime during the pandemic and the further limitations it is placing on the law enforcement community, as officers fall sick and even die, technologies such as AI are increasingly attractive. Yet, this pandemic equally serves to underscore the importance of adhering to the notions of responsible AI innovation as highlighted in the new report. Even in times of crisis, law enforcement must ensure that its use of AI is consistent with fundamental principles, rights and the rule of law.
Anita Hazenberg, Director of INTERPOL's Innovation Centre in Singapore, and Irakli Beridze, Head of UNICRI Centre for AI and Robotics, explored the many nuances of this matter at the Virtual Discussion Room, together with high level international experts from all regions of the world. The Virtual Discussion Room, which was the fifth in a series of Virtual Discussion Rooms organized by INTERPOL in response to the pandemic, was attended by more than 70 participants from the broader law enforcement and AI communities.