AI-Enabled First-Response Support After Sexual and Gender-Based Violence: A PRISMA-ScR Scoping Review

Area of Science:

• Digital Health
• Human-Computer Interaction
• Public Health

Background:

• Artificial intelligence (AI) is increasingly proposed to aid survivors of sexual and gender-based violence (GBV).
• Applications span crisis hotlines, legal tools, and healthcare pathways.
• The evidence base for AI in GBV support requires further investigation regarding scope, maturity, and evaluation.

Purpose of the Study:

• To map the application domains of AI in early support for GBV survivors.
• To assess the evaluative maturity of the existing evidence base.
• To identify implementation and safety gaps in AI-driven GBV support.

Main Methods:

• A scoping review was conducted following PRISMA-ScR guidelines.
• A Population-Concept-Context framework focused on AI-enabled first response and early support.
• Searches were performed across multiple databases, including specialized technical libraries.

Main Results:

• 27 sources of evidence were included, covering diverse AI applications like conversational support, triage, and legal routing.
• Most studies reported technical performance, usability, or acceptability.
• No workflow-integrated evaluations were identified; effectiveness, service uptake, and adverse events were rarely reported.

Conclusions:

• The evidence for AI in GBV support is heterogeneous and early-stage.
• Limited data exists on service-integrated effectiveness and safety.
• Key concerns include escalation, accountability, equity, data safety, and human handover, highlighting priorities for future research.