AI Agents Need Interfaces Beyond Chat
Direct editing, precise comments and durable records keep agent actions attached to the work they concern.
Developed from a conversation between Pete Winn and Andy David
Chat works for discussion and delegation, but it becomes awkward when every instruction must also describe where an action belongs. Documents and task boards already give work a visible structure. Letting people and AI agents act on those objects keeps the instruction, its target and the resulting change together instead of scattering them across a conversation.
Editing a document shows the difference. Someone can retype a passage and have the revision appear as a tracked change, or attach a comment to the exact sentence that needs attention. They don’t have to describe the sentence’s position in chat and hope the agent finds the right one. A task board provides the same orientation for ongoing work because each task has a defined place that people and agents can revisit.
Businesses will still need records even when agents perform more of the work. Documents preserve decisions and outputs beyond the immediate exchange, while structured interfaces make those records easier to inspect and change. Chat won’t disappear, just as messaging hasn’t replaced documents, spreadsheets or task lists. The useful shift is to let conversation initiate the work, then handle the work itself in the document or task where its context remains visible.