"There's so much responsibility on users right now:" Expert Advice for Staying Safer From Hate and Harassment

Title of the Paper

“There’s so much responsibility on users right now:” Expert Advice for Staying Safer From Hate and Harassment

Paper Information

• Subject Area: Digital Safety in Online Hate and Harassment
• Keywords: Safety and Privacy, Hate, Harassment, Online Advice, Data Security, Prevention Strategies, User Responsibility

Research Background and Issues

• Problems and Challenges:

  • Beyond specific groups (e.g., content creators, journalists, gamers), any internet user can become a target of hate and harassment.
  • Online hate and harassment can cause psychological, physical, and even financial harm through harmful content, involuntary information leaks, device surveillance, etc.
  • Existing safety advice is fragmented and often tailored to specific at-risk groups or threats, lacking general guidance applicable to all users.
  • Users inevitably bear the responsibility of preventing online attacks, increasing their psychological burden.

• Research Significance:

  • The increasing complexity of hate and harassment behaviors necessitates effective and universal prevention strategies.
  • General safety advice can help users who are uncertain about their risk level to enhance their awareness of online safety.

• Research Motivation and Related Work:

  • This paper aims to integrate existing resources and interview security experts to provide universally applicable advice.
  • It utilizes Thomas et al.'s framework for categorizing experiences of online hate and harassment, including harmful content, information leaks, device surveillance, impersonation, etc.

Solutions

• Methods and Solutions:

  • This study conducted interviews with 24 experts in the field of online hate and harassment to explore the feasibility of general digital safety advice.
  • A total of 219 existing safety recommendations were collected, deduplicated, and consolidated into 45 suggestions. The research team categorized these suggestions based on the types of threats they address.
  • Experts prioritized the recommendations based on the severity, prevalence of threats, and users' ability to self-protect.
  • The following research questions were explored:
    • Which threats should be prioritized?
    • What is the prioritization of existing safety recommendations?
    • Which overall safety strategies do experts believe are most helpful to users?

• Innovations:

  • Provided a prioritized evaluation report of digital safety recommendations for general users against hate and harassment.
  • Synthesized key strategies for preventing online harassment while revealing limitations in existing advice.

• Implementation Steps and Techniques:

  • Strategy Categorization: Recommendations were grouped into seven major threat categories, including harmful content, information leaks, device surveillance, etc.
  • Expert Prioritization: Feedback was collected using ranking tasks and card sorting techniques.
  • Analysis Methods: Combined qualitative and quantitative data analysis, supplemented by thematic analysis techniques.

Research Findings

• Specific Findings:

  • Experts unanimously agreed that users should prioritize protection against three types of threats: harmful content, information leaks, and device surveillance.
  • High-priority recommendations include enabling two-factor authentication, limiting the public exposure of personal information (e.g., home addresses and phone numbers), and using privacy tools (e.g., encrypted messaging applications).

• Advantages:

  • Optimized user-targeted recommendations, providing a clear and universally applicable guide compared to existing fragmented resources.
  • Minimal impact on user behavior while effectively reducing the risk of attacks.

• Experimental or Evaluation Results:

  • Experts ranked recommendations for each threat category, with high-priority suggestions focusing on account security, privacy control, and data minimization.
  • Certain threat types (e.g., impersonation, false reporting) were deemed difficult to defend against due to a lack of actionable recommendations.

• Limitations and Future Directions:

  • Limitations: The effectiveness of existing recommendations is limited to specific contexts, and additional work is needed to support high-risk individuals; ordinary users lack the capacity to address complex threats.
  • Future Directions:
    • Advocate for platforms to develop more robust security tools, such as automated community detection of hate behaviors.
    • Promote personalized recommendations based on user-specific risk models while ensuring the educational accessibility of general advice.
    • Foster a healthy digital community culture to reduce the burden of responsibility on individual users.

Through this study, the authors identified the primary security threats and countermeasures in the field of online hate and harassment, while also exposing the shortcomings of current systems. This provides a reference path for the design of future security tools and educational materials.