From "Fail Fast" to "Mature Safely:" Expert Perspectives as Secondary Stakeholders on Teen-Centered Social Media Risk Detection

• URL: http://arxiv.org/abs/2601.13516v1
• Date: Tue, 20 Jan 2026 02:09:54 GMT
• Title: From "Fail Fast" to "Mature Safely:" Expert Perspectives as Secondary Stakeholders on Teen-Centered Social Media Risk Detection
• Authors: Renkai Ma, Ashwaq Alsoubai, Jinkyung Katie Park, Pamela J. Wisniewski,
• Abstract summary: We present an evaluation of a teen-centered social media risk detection dashboard through online interviews with 33 online safety experts.<n>Experts praised our dashboard's clear design for teen agency, their feedback revealed five primary tensions in implementing and sustaining such technology.<n>These findings motivate us to rethink "teen-centered" and a shift from a "fail fast" to a "mature safely" paradigm for youth safety technology innovation.
• Score: 26.719055999377716
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: In addressing various risks on social media, the HCI community has advocated for teen-centered risk detection technologies over platform-based, parent-centered features. However, their real-world viability remains underexplored by secondary stakeholders beyond the family unit. Therefore, we present an evaluation of a teen-centered social media risk detection dashboard through online interviews with 33 online safety experts. While experts praised our dashboard's clear design for teen agency, their feedback revealed five primary tensions in implementing and sustaining such technology: objective vs. context-dependent risk definition, informing risks vs. meaningful intervention, teen empowerment vs. motivation, need for data vs. data privacy, and independence vs. sustainability. These findings motivate us to rethink "teen-centered" and a shift from a "fail fast" to a "mature safely" paradigm for youth safety technology innovation. We offer design implications for addressing these tensions before system deployment with teens and strategies for aligning secondary stakeholders' interests to deploy and sustain such technologies in the broader ecosystem of youth online safety.

Related papers

• A System of Care, Not Control: Co-Designing Online Safety and Wellbeing Solutions with Guardians ad Litem for Youth in Child Welfare [2.5142646517195693]
  We conducted a two-part workshop with 10 Guardians ad Litem (GALs) to explore their perspectives on online safety.<n>Our findings revealed that GALs struggle to support youth with online safety challenges due to limited digital literacy.<n>Gals emphasize developing tools that enable ongoing communication, therapeutic support, and coordination across stakeholders.
  arXiv  Detail & Related papers  (2026-02-15T04:56:26Z)
• CASTLE: A Comprehensive Benchmark for Evaluating Student-Tailored Personalized Safety in Large Language Models [55.0103764229311]
  We propose the concept of Student-Tailored Personalized Safety and construct CASTLE based on educational theories.<n>This benchmark covers 15 educational safety risks and 14 student attributes, comprising 92,908 bilingual scenarios.
  arXiv  Detail & Related papers  (2026-02-05T13:13:19Z)
• OS-Sentinel: Towards Safety-Enhanced Mobile GUI Agents via Hybrid Validation in Realistic Workflows [77.95511352806261]
  Computer-using agents powered by Vision-Language Models (VLMs) have demonstrated human-like capabilities in operating digital environments like mobile platforms.<n>We propose OS-Sentinel, a novel hybrid safety detection framework that combines a Formal Verifier for detecting explicit system-level violations with a Contextual Judge for assessing contextual risks and agent actions.
  arXiv  Detail & Related papers  (2025-10-28T13:22:39Z)
• Oyster-I: Beyond Refusal -- Constructive Safety Alignment for Responsible Language Models [93.5740266114488]
  Constructive Safety Alignment (CSA) protects against malicious misuse while actively guiding vulnerable users toward safe and helpful results.<n>Oy1 achieves state-of-the-art safety among open models while retaining high general capabilities.<n>We release Oy1, code, and the benchmark to support responsible, user-centered AI.
  arXiv  Detail & Related papers  (2025-09-02T03:04:27Z)
• SproutBench: A Benchmark for Safe and Ethical Large Language Models for Youth [14.569766143989531]
  The rapid proliferation of large language models (LLMs) in applications targeting children and adolescents necessitates a fundamental reassessment of prevailing AI safety frameworks.<n>This paper highlights key deficiencies in existing LLM safety benchmarks, including their inadequate coverage of age-specific cognitive, emotional, and social risks.<n>We introduce SproutBench, an innovative evaluation suite comprising 1,283 developmentally grounded adversarial prompts designed to probe risks such as emotional dependency, privacy violations, and imitation of hazardous behaviors.
  arXiv  Detail & Related papers  (2025-08-14T18:21:39Z)
• 'Teens Need to Be Educated on the Danger': Digital Access, Online Risks, and Safety Practices Among Nigerian Adolescents [0.0]
  This study investigates the online experiences of 409 secondary school adolescents in Nigeria's Federal Capital Territory (FCT)<n>Using self-administered surveys, we found that while most adolescents reported moderate access to online technology and connectivity, those who encountered risks frequently reported exposure to inappropriate content and online scams.<n>Parents emerged as the primary support network, though monitoring practices and communication varied widely.
  arXiv  Detail & Related papers  (2025-07-11T15:39:01Z)
• Safe-Child-LLM: A Developmental Benchmark for Evaluating LLM Safety in Child-LLM Interactions [8.018569128518187]
  We introduce Safe-Child-LLM, a benchmark and dataset for assessing AI safety across two developmental stages: children (7-12) and adolescents (13-17).<n>Our framework includes a novel multi-part dataset of 200 adversarial prompts, curated from red-teaming corpora, with human-annotated labels for jailbreak success and a standardized 0-5 ethical refusal scale.<n> evaluating leading LLMs -- including ChatGPT, Claude, Gemini, LLaMA, DeepSeek, Grok, Vicuna, and Mistral -- we uncover critical safety deficiencies in child-facing scenarios.
  arXiv  Detail & Related papers  (2025-06-16T14:04:54Z)
• Towards Resilience and Autonomy-based Approaches for Adolescents Online Safety [7.446834742371107]
  We discuss the paradigm shift that has emerged in the literature to move toward resilient-based and privacy-preserving solutions to promote adolescents' online safety.<n>We highlight the limitations of restrictive mediation strategies, which often induce a trade-off between teens' privacy and online safety.<n>We present an overview of empirical studies that conceptualized and examined resilience-based approaches to promoting the digital well-being of teens.
  arXiv  Detail & Related papers  (2025-04-22T02:23:48Z)
• Building a Village: A Multi-stakeholder Approach to Open Innovation and Shared Governance to Promote Youth Online Safety [13.270849736222065]
  We interviewed 33 stakeholders in the space of youth online safety to understand where their visions toward working together converged and surfaced tensions.<n>We found that non-partisan leadership is necessary to chart actionable, equitable goals to facilitate collaboration between stakeholders.<n>We propose the creation of an open-innovation organization that unifies the diverse voices in youth online safety to develop open-standards and evidence-based design patterns.
  arXiv  Detail & Related papers  (2025-04-04T22:26:37Z)
• AILuminate: Introducing v1.0 of the AI Risk and Reliability Benchmark from MLCommons [62.374792825813394]
  This paper introduces AILuminate v1.0, the first comprehensive industry-standard benchmark for assessing AI-product risk and reliability.<n>The benchmark evaluates an AI system's resistance to prompts designed to elicit dangerous, illegal, or undesirable behavior in 12 hazard categories.
  arXiv  Detail & Related papers  (2025-02-19T05:58:52Z)
• Foveate, Attribute, and Rationalize: Towards Physically Safe and Trustworthy AI [76.28956947107372]
  Covertly unsafe text is an area of particular interest, as such text may arise from everyday scenarios and are challenging to detect as harmful. We propose FARM, a novel framework leveraging external knowledge for trustworthy rationale generation in the context of safety. Our experiments show that FARM obtains state-of-the-art results on the SafeText dataset, showing absolute improvement in safety classification accuracy by 5.9%.
  arXiv  Detail & Related papers  (2022-12-19T17:51:47Z)
• Is 40 the new 60? How popular media portrays the employability of older software developers [78.42660996736939]
  We analyzed popular online articles and related discussions on Hacker News through the lens of employability issues and potential mitigation strategies. We highlight the importance of keeping up-to-date, specializing in certain tasks or technologies, and present role transitions as a way forward for veteran developers.
  arXiv  Detail & Related papers  (2020-04-13T10:00:03Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.