Long-horizon agents, which can run for hours or days, require a dual-mode UI. Users need an asynchronous way to manage multiple running agents (like a Jira board or inbox). However, they also need to seamlessly switch to a synchronous chat interface to provide real-time feedback or corrections when an agent pauses or finishes.
Related Insights
Unlike standard chatbots where you wait for a response before proceeding, Cowork allows users to assign long-running tasks and queue new requests while the AI is working. This shifts the interaction from a turn-by-turn conversation to a delegated task model.
Engineer productivity with AI agents hits a "valley of death" at medium autonomy. The tools excel at highly responsive, quick tasks (low autonomy) and fully delegated background jobs (high autonomy). The frustrating middle ground is where it's "not enough to delegate and not fun to wait," creating a key UX challenge.
The primary interface for managing AI agents won't be simple chat, but sophisticated IDE-like environments for all knowledge workers. This paradigm of "macro delegation, micro-steering" will create new software categories like the "accountant IDE" or "lawyer IDE" for orchestrating complex AI work.
While chatbots are an effective entry point, they are limiting for complex creative tasks. The next wave of AI products will feature specialized user interfaces that combine fine-grained, gesture-based controls for professionals with hands-off automation for simpler tasks.
As AI moves into collaborative 'multiplayer mode,' its user interface will evolve into a command center. This UI will explicitly separate tasks agents can execute autonomously from those requiring human intervention, which are flagged for review. This shifts the user's role from performing tasks to overseeing and approving AI's work.
The best UI for an AI tool is a direct function of the underlying model's power. A more capable model unlocks more autonomous 'form factors.' For example, the sudden rise of CLI agents was only possible once models like Claude 3 became capable enough to reliably handle multi-step tasks.
The evolution from AI autocomplete to chat is reaching its next phase: parallel agents. Replit's CEO Amjad Masad argues the next major productivity gain will come not from a single, better agent, but from environments where a developer manages tens of agents working simultaneously on different features.
While chat works for human-AI interaction, the infinite canvas is a superior paradigm for multi-agent and human-AI collaboration. It allows for simultaneous, non-distracting parallel work, asynchronous handoffs, and persistent spatial context—all of which are difficult to achieve in a linear, turn-based chat interface.
The ideal AI-powered engineering workflow isn't just one tool, but a fluid cycle. It involves synchronous collaboration with an AI for planning and review, then handing off to an asynchronous agent for implementation and testing, before returning to synchronous mode for the next phase.
As Siri integrates powerful LLMs like Gemini, a simple voice interface is insufficient. A dedicated app is necessary for users to review conversation history and interact asynchronously, much like texting a human assistant, to handle complex, multi-turn interactions.
