Adobe CX Enterprise Coworker Chat Overview overview
Coworker Chat is a conversational interface for getting customer experience work done. You describe a goal in plain language (e.g., “build an audience for the spring yoga promotion,” or, “tell me why this audience shrank last week”) and Coworker Chat plans the work, executes it across your Adobe and connected systems, validates the results, and brings the finished work back to you for approval.
The value: work that used to span multiple screens, teams, and days now happens in one conversation in minutes. You delegate the outcome and stay in control through review and approval; Coworker Chat does the steps, recovers from errors along the way, and gets sharper as it learns how your team works.
An evolution from AI Assistant
If you have used AI Assistant and Adobe Experience Platform Agent Orchestrator, Coworker Chat is the next progression. AI Assistant is evolving into Coworker Chat. Everything you can do in AI Assistant today by invoking Adobe purpose-built Agents (product knowledge, operational and data insights, the conversational experience) will remain available. Coworker Chat is backed by the Coworker building blocks (the enterprise harness, skills, governance, and memory), so the same conversation can now go beyond answering and actually complete the work end to end.
The chat experience
- Rich, in-conversation results: Charts, tables, and interactive artifacts render inline, right where you are working. You can upload an image for Coworker to use, and preview PDFs in the conversation, with no switching between tools.
- It shows its work: As Coworker Chat takes on a task, you can watch it reason and progress (loading the right skill, checking your data model, running a compliance check, looking for overlaps), so you always know what it is doing and why.
- Long, complex sessions: Coworker Chat sustains extended, multi-step sessions and holds context across them, so you can work a problem from start to finish in one place.
- Suggested and follow-up prompts: Coworker Chat proposes next steps, so you do not have to guess what to ask.
- Conversation history: Past conversations are saved, so you can return to earlier work, pick up where you left off, and reuse what Coworker Chat produced.
- Feedback built in: Rate any response with a thumbs up or thumbs down to tell Coworker Chat what worked and what did not; your feedback helps improve the experience over time.
Building blocks
Coworker Chat is powered by the Coworker building blocks. The following building blocks are available today.
- Enterprise harness: the engine that turns your goal into completed work. It runs a continuous, self-correcting loop (reason, select a skill, execute, validate, repeat), recovering from errors and working in parallel until the job is done, not just attempted.
- Skills: reusable playbooks, in plain Markdown, for doing each task well. Coworker Chat ships with skills for the most common customer experience tasks today, and the goal is to cover the work across all Adobe CX Enterprise applications over time.
- MCP connectors: governed access to act across Adobe and third-party systems, with out-of-the-box connectors and the option to bring your own.
- Governance: approval gates, permissions inherited from your Adobe organization, checks that run before an action executes, and a full audit trail.
- Memory: Coworker Chat learns your preferences, decisions, and corrections across sessions, scoped from organization-wide rules down to your personal preferences, so it gets sharper the more you work with it.