Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

The Scenic Route to Deception: Dark Patterns and Explainability Pitfalls in Conversational Navigation

Ilya Ilyankou, Stefano Cavazzi, James Haworth

Abstract

As pedestrian navigation increasingly experiments with Generative AI, and in particular Large Language Models, the nature of routing risks transforming from a verifiable geometric task into an opaque, persuasive dialogue. While conversational interfaces promise personalisation, they introduce risks of manipulation and misplaced trust. We categorise these risks using a 2x2 framework based on intent and origin, distinguishing between intentional manipulations (dark patterns) and unintended harms (explainability pitfalls). We propose seamful design strategies to mitigate these harms. We suggest that one robust way to operationalise trustworthy conversational navigation is through neuro-symbolic architecture, where verifiable pathfinding algorithms ground GenAI's persuasive capabilities, ensuring systems explain their limitations and incentives as clearly as they explain the route.

The Scenic Route to Deception: Dark Patterns and Explainability Pitfalls in Conversational Navigation

Abstract

As pedestrian navigation increasingly experiments with Generative AI, and in particular Large Language Models, the nature of routing risks transforming from a verifiable geometric task into an opaque, persuasive dialogue. While conversational interfaces promise personalisation, they introduce risks of manipulation and misplaced trust. We categorise these risks using a 2x2 framework based on intent and origin, distinguishing between intentional manipulations (dark patterns) and unintended harms (explainability pitfalls). We propose seamful design strategies to mitigate these harms. We suggest that one robust way to operationalise trustworthy conversational navigation is through neuro-symbolic architecture, where verifiable pathfinding algorithms ground GenAI's persuasive capabilities, ensuring systems explain their limitations and incentives as clearly as they explain the route.
Paper Structure (16 sections, 1 figure, 1 table)

This paper contains 16 sections, 1 figure, 1 table.

Table of Contents

  1. Introduction
  2. Related work
  3. Categorising Harms: A $2\times2$ Framework
  4. The harm
  5. The harm
  6. The harm
  7. Towards 'Honest' Navigation: Seamful Design and Calibrated Trust
  8. Sponsorship Disclosure
  9. Trust calibration
  10. Progressive Consent
  11. Technical Feasibility
  12. Reference Implementation
  13. Operationalising the DP/EP Distinction via Route Comparison
  14. Limitations
  15. Conclusion
  16. ...and 1 more sections

Figures (1)

  • Figure 1: Conceptual Architecture of a secure, auditable AI workflow for a conversational navigation system following the proposed seamful design and calibrated trust. The process begins with the User interacting through an LLM Interpreter, which processes the prompt before passing it to the Symbolic Routing Engine, which directs requests to an LLM Verbaliser for language generation. Supporting layers include Telemetry & Provenance and Audit Log for traceability, connected to Compute and Storage resources. The entire system is governed by Versioning, Auditability, and Policy Enforcement to ensure compliance and accountability.