Yoti Facial Age Estimation Technology White Paper

This white paper presents Yoti's facial age estimation technology, which determines a person's age using an image of their face without requiring physical documentation or human intervention. The technology is designed to be accurate across various skin types and genders, ensuring that businesses can rely on estimated ages as valid proof of age. The model adheres to GDPR principles, emphasizing privacy by design, as it cannot identify individuals and only requires a facial image, which is promptly deleted after use. The paper details the technology's performance metrics, including a True Positive Rate (TPR) of 99.93% for correctly estimating ages of 13-17 year-olds as under 25, and 98.35% for 6-11 year-olds as under 13. It also discusses the ethical considerations in data collection and training, highlighting efforts to minimize bias across different demographics. Additionally, the paper outlines ongoing improvements in accuracy and the importance of compliance with regulatory standards.