Typed prompts require translating intent into text, which can lose information at each step for visual or spatial tasks. Earlier interaction design moved toward direct manipulation such as seeing, pointing, and dragging to avoid that kind of translation strain. A key question is why this mismatch was not felt as sharply in 2022 and why it is now becoming more noticeable. The framing uses a vocabulary centered on “nodes,” describing communication between high-dimensional minds and the asymmetry that emerges when one side must compress rich perception into text while the other side operates on different internal representations. The change in felt strain is treated as a signal about what current AI systems ask from users.
"On AI, prompts, and the strange new asymmetry between two high-dimensional minds"
Read at Medium
Unable to calculate read time
Collection
[
|
...
]