Researcher warns against one-size-fits-all age verification online

admin

Official Verified Account

Admin

posted on 1 month ago — updated on 1 second ago

3
views

A comprehensive analysis of age assurance technologies co-authored by CISPA faculty member Dr. Wouter Lueks concludes that many currently proposed systems for protecting minors online come with significant trade-offs for privacy, security, accessibility and civil liberties. The study "Assessing Age Assurance Technologies: Effectiveness, Side-Effects, and Acceptance," conducted with colleagues from the University of Hamburg and the Leibniz Institute for Media Research, evaluates a broad range of approaches currently being debated by policymakers and platforms worldwide.

July 20, 2026

by Felix Koltermann, Helmholtz Association of German Research Centres

edited by Swati Mestri, reviewed by Andrew Zinin

Scientific Editor

Chief Editor

This article has been reviewed according to Science X's editorial process and policies. Editors have highlighted the following attributes while ensuring the content's credibility:

fact-checked

preprint

trusted source

proofread

Add as preferred source

parental control on devices Credit: Atlantic Ambience from Pexels

Age assurance technologies (AATs) have recently attracted significant public attention in the debate around online child protection. "There is a belief in society that we need to protect children from online harms," Lueks explains. "While this debate used to focus mostly on age-restricted content such as gambling and pornography, it has expanded to include a broader range of services, including social media services and other forms of potentially harmful content."

Against this backdrop, Lueks and his colleagues analyzed different technological architectures and regulatory approaches emerging across Europe, the United Kingdom and other jurisdictions. While some of the analyzed approaches can provide a high level of security, the study finds that stronger forms of verification often come with greater societal costs.

In the paper, the authors distinguish four different approaches to age assurance: parental control and consent on devices, age estimation, age inference and age verification. In the first approach, parents configure a child's device to block or approve access to specific apps or websites.

Lueks explains that "age inference uses systems to analyze behavioral patterns, possibly over time, to infer a user's age." In contrast, "age estimation uses systems to analyze a selfie or other information to estimate someone's age." Age verification, meanwhile, relies on official documents or trusted sources to determine a user's age.

The researchers find that these approaches differ significantly not only in how reliably they can determine a user's age but also in their implications for privacy, security, accessibility and fundamental rights.

The findings suggest that stronger age-assurance measures often come with broader societal burdens. Depending on the approach, these may include the collection and storage of sensitive personal or biometric data, increased tracking and profiling of users, the exclusion of people without suitable identity documents and biased or inaccurate assessments.

Age inference systems in particular, which analyze online behavior to determine a user's age, risk normalizing large-scale surveillance while providing only limited assurance. Moreover, with the exception of parental control and consent, none of these approaches effectively prevents circumvention through the use of VPNs.

As a result, no single solution emerges as universally preferable across all use cases. A broader concern is the potential for censorship. "Right now, the stated goal is to protect children from harmful content, such as social media," Lueks explains. "Tomorrow, the same systems could be used to restrict access to online games, LGBT content, political material or books."

Based on their assessment, Lueks and his colleagues urge policymakers to carefully distinguish between age assurance systems rather than treating them as equally acceptable or equally effective. Instead, they propose a graduated assessment framework that weighs effectiveness against societal side effects and impacts on fundamental rights.

"Because these systems affect everyone and are relatively easy to circumvent, we argue for lighter-weight solutions," Lueks says. "Our first preference is parental control and consent on devices." This may not be perfect, but it has fewer negative externalities because it does not affect everyone universally.

"A second possibility is privacy-preserving on-device age verification as a second line of defense," he continues. "Our third preference is credential-based wallet solutions with strong privacy guarantees."

Lueks emphasizes that his goal is not to reject child protection measures in general: "Protecting minors online is essential. But measures must be proportionate, privacy-preserving and realistically deployable at scale without creating new harms to the whole population."

The authors stress that technical measures alone are unlikely to fully solve the underlying societal challenges surrounding minors' online experiences. According to the researchers, effective child protection will require a broader combination of technological safeguards, platform responsibility, parental support, education and public policy. The paper enriches the debate by providing a fundamental analysis of the advantages and disadvantages of current technological solutions.

Wouter Lueks et al, Assessing Age Assurance Technologies: Effectiveness, Side-Effects, and Acceptance, arXiv (2026). DOI: 10.48550/arxiv.2603.25695

Journal information: arXiv

Swati Mestri holds a bachelor's degree in Electronics Engineering and has worked as a content editor since 2019. She has experience editing research documents across technology, health care, and materials science, and has a particular interest in technology and space. Full profile →

Master's in physics with research experience. Long-time science news enthusiast. Plays key role in Science X's editorial success. Full profile →

Jul 21, 2026

20 hours ago

Jul 20, 2026

4 hours ago

13 hours ago

17 hours ago

18 hours ago

19 hours ago

20 hours ago

21 hours ago

Oct 28, 2024