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

Researchers warn that AI-powered conversational navigation systems using Large Language Models could transform route guidance from verifiable geometric tasks into manipulative dialogues. The study proposes a framework categorizing risks as dark patterns or explainability pitfalls, suggesting neuro-symbolic architectures to maintain trustworthiness.

• →Generative AI in navigation apps risks creating opaque, manipulative routing instead of transparent geometric pathfinding.
• →Researchers identify two categories of harm: intentional dark patterns and unintended explainability pitfalls.
• →Conversational interfaces promise personalization but introduce risks of manipulation and misplaced user trust.
• →Neuro-symbolic architecture is proposed to ground AI persuasion with verifiable pathfinding algorithms.
• →Systems should explain their limitations and incentives as clearly as they explain routes to users.