Yes. Assistive technologies (screen readers, voice control, switch devices, accessibility browser extensions, etc.) can change several contextual signals that RBA systems use, and so they can affect risk scoring.
Key ways assistive tech can influence RBA:
- Device and browser fingerprints: Accessibility tools may alter user-agent strings, installed fonts, or accessibility-related browser APIs, producing fingerprints that differ from typical profiles.
- Interaction patterns: Users of assistive tech often have different timing, click/typing rhythms, and navigation flows that RBA behavioral models might flag as anomalous.
- Device reputation and sensors: External adaptive hardware (e.g., specialized input devices) or virtualized environments can change device IDs, sensor outputs, or other telemetry.
- Location and network signals: Some assistive services route traffic differently (proxies, remote desktops, cloud-based screen readers), which can alter IP, geolocation, or latency signals.
- Accessibility extensions and privacy tools: These can block or modify headers, cookies, or tracking scripts that RBA relies on.
Implications and best practices:
- Higher false positives: Legitimate users of assistive tech may face more step-ups or blocks unless the RBA is tuned to expect such variation.
- Inclusive baselines: Build behavioral baselines that include accessibility-typical patterns and allow users to register known assistive setups.
- Transparent fallback options: Provide clear, accessible step-up methods (e.g., accessible OTP delivery) and account recovery paths.
- Privacy and consent: Be cautious collecting sensitive accessibility-related data; follow privacy law and avoid inferring disability without consent.
References:
- OWASP, “Risk-Based Authentication Cheat Sheet.”
- NIST SP 800-63B, Digital Identity Guidelines (authentication considerations).