The World Economic Forum, in partnership with the International Criminal Police Organization (INTERPOL), the United Nations Interregional Crime and Justice Research Institute (UNICRI), and the Netherlands police, published A Policy Framework for Responsible Limits on Facial Recognition Technology, Use Case: Law Enforcement Investigations.
While the adoption of facial recognition technology (FRT) has many socially beneficial uses in various industries, it also creates a unique set of challenges that require an appropriate governance process to ensure its ethical and human rights compliant use. To address these challenges, the World Economic Forum and project community published its initial vision for a policy framework for the responsible use of FRT in law enforcement investigations in 2021.
Following its launch, six law enforcement agencies undertook an exercise to pilot the policy framework in order to review and validate its utility and completeness. Drawing from the feedback and learnings of the pilots, the new insight report highlights key principles such as respect for human and fundamental rights, necessary and proportional use, mitigation of error and bias, and transparency. It also includes a self-assessment questionnaire used to support law enforcement agencies in effectively implementing these the proposed principles for responsible facial recognition use. Through a series of questions, it prompts agencies to reflect on their governance systems and their use of FRT in terms of the stated principles.
“This report shows the depth and breadth of how the Forum is able to bring multiple stakeholders together to find solutions to the most difficult technological challenges of our time. We are delighted to have created a document which is already being adopted by others for their governance regimes and by our Partners in the work,” said Kay Firth-Butterfield, Head of AI and Machine Learning at the World Economic Forum. “Now we encourage others around the world to adopt regulations or best practices reflecting this work.”
The International Organization for Standardization – ISO/IEC JTC1 SC37 – is currently developing a standard titled “Biometric identification systems involving passive capture subjects.” The endeavour is intended to support the implementation of the proposed European Union (EU) AI Act. The standard will provide guidance and requirements for providers of biometric systems labelled as “high risks: in the draft EU AI Act and will include some of the recommendations from the framework on the responsible use of facial recognition technology for law enforcement investigations developed by the World Economic Forum, INTERPOL, UNICRI and the Netherlands Police.
“We have co-designed this framework to serve as a unique reference to law enforcement in our 195 member countries on the responsible and transparent use of facial recognition”, said Cyril Gout, Director of Operational Support and Analysis, INTERPOL. “We will support its implementation through our global police network to increase awareness of this important biometric technology. More than 2,000 terrorists, criminals, fugitives, persons of interest or missing persons have been identified since the launch of INTERPOL’s facial recognition system in 2016.”
A total of six law enforcement agencies from five countries participated in the piloting the framework: the Brazilian Federal Police, the Central Directorate of the Judicial Police in France, the National Gendarmerie in France, the Netherlands Police, the New Zealand Police, and the Swedish Police Authority. These agencies deliberated upon the policy framework for four months before completing the self-assessment questionnaire and providing their feedback of its utility and completeness.
“We are particularly proud of the policy framework for the use of facial recognition technology and are confident it will help law enforcement agencies to ensure their use of this technology respects fundamental and human rights. As it now goes out into the world, we and our partners in the Forum, INTERPOL and the Netherlands Police stand ready to support law enforcement agencies to implement its principles. This can also be a valuable resource for the general public and we encourage all those interested in, or concerned by, this technology to reflect on it.” said Irakli Beridze, Head, UNICRI Centre for Artificial Intelligence and Robotics.
As the framework now enters the next phase of its lifecycle, the project team encourages those engaged in the global policy debate about the governance of facial recognition technology to join in these efforts and to promote the adoption and deployment of governance frameworks such as this.