Guide-by-Pointing

A prompting technique for multimodal AI assistants where a user extends their hand into the camera's field of view and asks the AI to identify what they are pointing at, or to provide spatial directions for moving their hand toward a specific item. This technique enables blind users to physically locate and interact with objects in their environment using AI-guided spatial feedback. Current limitations include unreliable directional cues and weak egocentric spatial reasoning by the AI models.