Age checks are becoming increasingly common worldwide, with over 30 age assurance laws in force globally. As governments crack down on online safety, platforms must ensure they can verify users’ ages without compromising their data. One solution is emerging as the new standard: on-device age estimation.
The Problem with Server-Based Age Estimation
Currently, most age verification systems transmit users’ faces to a server for analysis. However, this approach has several drawbacks. Firstly, it leaves sensitive biometric data vulnerable to interception and breaches. According to the Identity Theft Resource Center’s 2025 Annual Data Breach Report, the U.S. recorded a record high of 3,322 data compromises last year, with supply-chain breaches doubling over the same period.
Secondly, server-based age estimation is susceptible to agentic fraud – attempts carried out using AI agents. Incode Technologies has tracked this rise in agentic fraud, which made up 40% of fraud attempts by the first quarter of 2026 and is estimated to exceed 90% within the next 18 months.
On-Device Age Estimation: A New Standard for Age Verification
Incode’s facial age estimation and passive liveness models now run entirely on the user’s device, ensuring that their face never leaves the device. This approach eliminates the need for server-based processing, reducing the risk of data breaches and agentic fraud.
The technology uses knowledge distillation to shrink Incode’s proprietary models to roughly a tenth of their original size, allowing them to run inside an ordinary browser or app without requiring special hardware.
Privacy by Architecture: A Shift in Approach
Traditional age verification methods rely on privacy policies, which are essentially written promises that biometric data will be handled with care and deleted after the check. However, these policies cannot stop breaches, insider threats, or compromised vendors; they can only assign responsibility afterward.
Incode’s approach is based on ‘privacy by architecture,’ where the system is designed so sensitive data never becomes accessible in the first place. By processing age estimation locally, users’ faces are not transmitted or stored, making it impossible for Incode or any client platform to access biometric data.
The Future of Age Verification: On-Device Solutions
Incode Technologies has made significant strides in advancing on-device age verification with its $100 million commitment to developing privacy-preserving identity infrastructure. Its latest product, On-Device Age Estimation, launched in July, marks a major milestone in this journey.
The future of age verification is clear: on-device solutions that prioritize user data protection and prevent agentic fraud will become the new standard. As governments continue to enforce stricter regulations, platforms must adapt to these changing requirements and invest in technologies that safeguard users’ sensitive information.
Source: Original article